Verdict Canonical Public Benchmark Suite

Audited Multi-Judge Evaluation Panel

Frontier AI Model Leaderboard (Mid-2026)

Evaluating 100+ frontier AI models on real-world "vibe coding" tasks, agentic repository refactoring, multimodal vision understanding, and real-time voice synthesis.

Intelligence

Claude Opus 5 (Adaptive Reasoning, Max Effort) & Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)lead composite reasoning & coding accuracy.

SOTA S-Tier Rankings

Output Speed

Celeris-1 & Mercury 2 top generation throughput (>2158 t/s).

Ultra Low Throughput Latency

Latency (TTFT)

DeepSeek V4 Flash 0731 (Reasoning, Max Effort) & MiniMax-M3 yield fastest first token response (<1ms).

Realtime Voice & Chat

Cost per Task

Gemma 4 E4B (Reasoning) & Gemma 4 E4B (Non-reasoning) offer SOTA intelligence at $0.02/1M input tokens.

Lowest Cost per Sample

Context Window

Llama 4 Scout & Grok 4.20 0309 v2 (Reasoning) support up to 10M token context windows.

Whole Repository Analysis

Artificial Analysis Index Rankings

Auto-Sync Active (artificialanalysis.ai)

Canonical AI Coding Model Rankings (578 Models)

Auto-Sync Active

Claude Opus 5 (Adaptive Reasoning, Max Effort)

Anthropic

60.7

Speed: 56 t/sContext: 1M

Details

Claude Opus 5 (Adaptive Reasoning, Xhigh Effort)

Anthropic

60.1

Speed: 57 t/sContext: 1M

Details

Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback)

Anthropic

59.9

Speed: 73 t/sContext: 1M

Details

GPT-5.6 Sol (max)

OpenAI

58.9

Speed: 71 t/sContext: 1M

Details

Claude Opus 5 (Adaptive Reasoning, High Effort)

Anthropic

58.9

Speed: 54 t/sContext: 1M

Details

GPT-5.6 Sol (xhigh)

OpenAI

57.7

Speed: 72 t/sContext: 1M

Details

Kimi K3 (max)Open

Kimi

57.1

Speed: 36 t/sContext: 1M

Details

Claude Opus 5 (Adaptive Reasoning, Medium Effort)

Anthropic

56.3

Speed: 54 t/sContext: 1M

Details

GPT-5.6 Sol (high)

OpenAI

55.9

Speed: 67 t/sContext: 1M

Details

#10

Claude Opus 4.8 (Adaptive Reasoning, Max Effort)

Anthropic

55.7

Speed: 58 t/sContext: 1M

Details

#11

GPT-5.6 Terra (max)

OpenAI

55.0

Speed: 143 t/sContext: 1M

Details

#12

GPT-5.5 (xhigh)

OpenAI

54.8

Speed: 87 t/sContext: 922k

Details

#13

Grok 4.5 (high)

SpaceXAI

53.8

Speed: 61 t/sContext: 500k

Details

#14

GPT-5.6 Sol (medium)

OpenAI

53.6

Speed: 64 t/sContext: 1M

Details

#15

Claude Opus 4.7 (Adaptive Reasoning, Max Effort)

Anthropic

53.5

Speed: 52 t/sContext: 1M

Details

#16

Claude Sonnet 5 (Adaptive Reasoning, Max Effort)

Anthropic

53.4

Speed: 81 t/sContext: 1M

Details

#17

GPT-5.5 (high)

OpenAI

53.1

Speed: 78 t/sContext: 922k

Details

#18

GPT-5.6 Terra (xhigh)

OpenAI

51.6

Speed: 124 t/sContext: 1M

Details

#19

GPT-5.4 (xhigh)

OpenAI

51.4

Speed: 143 t/sContext: 1.1M

Details

#20

GPT-5.6 Luna (max)

OpenAI

51.2

Speed: 181 t/sContext: 1M

Details

#21

GLM-5.2 (max)Open

Z AI

51.1

Speed: 186 t/sContext: 1M

Details

#22

Muse Spark 1.1 (xhigh)

Meta

50.6

Speed: 216 t/sContext: 1M

Details

#23

Claude Opus 5 (Adaptive Reasoning, Low Effort)

Anthropic

50.6

Speed: 56 t/sContext: 1M

Details

#24

GPT-5.5 (medium)

OpenAI

50.4

Speed: 74 t/sContext: 922k

Details

#25

Gemini 3.5 Flash (high)

Google

50.2

Speed: 187 t/sContext: 1M

Details

#26

Gemini 3.6 Flash (high)

Google

50.1

Speed: 213 t/sContext: 1M

Details

#27

DeepSeek V4 Flash 0731 (Reasoning, Max Effort)Open

DeepSeek

49.9

Speed: 124 t/sContext: 1M

Details

#28

GPT-5.6 Sol (low)

OpenAI

49.4

Speed: 65 t/sContext: 1M

Details

#29

GPT-5.6 Luna (xhigh)

OpenAI

49.1

Speed: 163 t/sContext: 1M

Details

#30

GPT-5.6 Terra (high)

OpenAI

49.0

Speed: 122 t/sContext: 1M

Details

#31

Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort)

Anthropic

47.2

Speed: 56 t/sContext: 1M

Details

#32

Kimi K3 (low)Open

Kimi

46.6

Speed: 36 t/sContext: 1M

Details

#33

Gemini 3.1 Pro Preview

Google

46.5

Speed: 129 t/sContext: 1M

Details

#34

GPT-5.6 Luna (high)

OpenAI

46.1

Speed: 166 t/sContext: 1M

Details

#35

Qwen3.7 Max

Alibaba

46.0

Speed: 205 t/sContext: 1M

Details

#36

GPT-5.6 Terra (medium)

OpenAI

45.6

Speed: 120 t/sContext: 1M

Details

#37

Gemini 3.5 Flash (medium)

Google

45.4

Speed: 188 t/sContext: 1M

Details

#38

MiniMax-M3Open

MiniMax

44.4

Speed: 92 t/sContext: 1M

Details

#39

DeepSeek V4 Pro (Reasoning, Max Effort)Open

DeepSeek

44.3

Speed: 66 t/sContext: 1M

Details

#40

GPT-5.3 Codex (xhigh)

OpenAI

44.3

Speed: 130 t/sContext: 400k

Details

#41

Kimi K2.6Open

Kimi

44.2

Speed: 38 t/sContext: 256k

Details

#42

Motif 3 (Beta)

Motif Technologies

44.1

Speed: —Context: 262k

Details

#43

Claude Opus 4.6 (Adaptive Reasoning, Max Effort)

Anthropic

43.7

Speed: 42 t/sContext: 1M

Details

#44

GPT-5.5 (low)

OpenAI

43.5

Speed: 72 t/sContext: 922k

Details

#45

DeepSeek V4 Pro (Reasoning, High Effort)Open

DeepSeek

43.1

Speed: 66 t/sContext: 1M

Details

#46

Muse Spark

Meta

43.1

Speed: —Context: 262k

Details

#47

Claude Opus 4.7 (Non-reasoning, High Effort)

Anthropic

42.7

Speed: 47 t/sContext: 1M

Details

#48

MiMo-V2.5-ProOpen

Xiaomi

42.2

Speed: 48 t/sContext: 1M

Details

#49

GPT-5.2 (xhigh)

OpenAI

42.2

Speed: 90 t/sContext: 400k

Details

#50

Kimi K2.7 CodeOpen

Kimi

41.9

Speed: 40 t/sContext: 256k

Details

#51

Claude Sonnet 5 (Non-reasoning, High Effort)

Anthropic

41.7

Speed: 65 t/sContext: 1M

Details

#52

Hy3Open

Tencent

41.2

Speed: 70 t/sContext: 256k

Details

#53

GPT-5.6 Sol (Non-reasoning)

OpenAI

41.2

Speed: 64 t/sContext: 1M

Details

#54

Nex-N2-ProOpen

Nex AGI

41.0

Speed: 130 t/sContext: 262k

Details

#55

Claude Opus 4.5 (Reasoning)

Anthropic

40.8

Speed: 49 t/sContext: 200k

Details

#56

Inkling (xhigh)Open

Thinking Machines

40.7

Speed: 85 t/sContext: 1M

Details

#57

GPT-5.6 Terra (low)

OpenAI

40.5

Speed: 122 t/sContext: 1M

Details

#58

MiMo-V2-Pro

Xiaomi

40.3

Speed: —Context: 1M

Details

#59

DeepSeek V4 Flash (Reasoning, Max Effort)Open

DeepSeek

40.3

Speed: —Context: 1M

Details

#60

Inkling SmallOpen

Thinking Machines

40.2

Speed: 132 t/sContext: 1M

Details

#61

GLM-5.1 (Reasoning)Open

Z AI

40.2

Speed: 73 t/sContext: 200k

Details

#62

GPT-5.2 Codex (xhigh)

OpenAI

40.1

Speed: —Context: 400k

Details

#63

Qwen3.6 Max Preview

Alibaba

40.0

Speed: 51 t/sContext: 256k

Details

#64

GPT-5.4 mini (xhigh)

OpenAI

40.0

Speed: 190 t/sContext: 400k

Details

#65

Grok Build 0.1 0616

SpaceXAI

39.8

Speed: 94 t/sContext: 256k

Details

#66

Qwen3.6 Plus

Alibaba

39.6

Speed: 56 t/sContext: 1M

Details

#67

Gemini 3 Pro Preview (high)

Google

39.6

Speed: —Context: 1M

Details

#68

GLM-5 (Reasoning)Open

Z AI

39.5

Speed: 40 t/sContext: 200k

Details

#69

GPT-5.4 (low)

OpenAI

39.1

Speed: 111 t/sContext: 1.1M

Details

#70

Qwen3.7 Plus

Alibaba

39.0

Speed: 53 t/sContext: 1M

Details

#71

JT-4.1 Flash 236B A21B

China Mobile

38.8

Speed: —Context: 256k

Details

#72

Agnes 2.5 Pro Alpha

Sapiens AI

38.8

Speed: 131 t/sContext: 1M

Details

#73

GPT-5.4 nano (xhigh)

OpenAI

38.2

Speed: 167 t/sContext: 400k

Details

#74

MiniMax-M2.7Open

MiniMax

38.1

Speed: 106 t/sContext: 205k

Details

#75

GLM-5-Turbo

Z AI

38.1

Speed: —Context: 200k

Details

#76

GPT-5.6 Luna (medium)

OpenAI

38.1

Speed: 156 t/sContext: 1M

Details

#77

GPT-5.2 (medium)

OpenAI

38.0

Speed: —Context: 400k

Details

#78

Claude Opus 4.6 (Non-reasoning, High Effort)

Anthropic

37.8

Speed: 37 t/sContext: 1M

Details

#79

Nemotron 3 Ultra 550B A55B (Reasoning)Open

NVIDIA

37.8

Speed: 135 t/sContext: 262k

Details

#80

Gemini 3 Flash Preview (Reasoning)

Google

37.8

Speed: 179 t/sContext: 1M

Details

#81

Grok 4.3 (high)

SpaceXAI

37.6

Speed: 141 t/sContext: 1M

Details

#82

DeepSeek V4 Flash (Reasoning, High Effort)Open

DeepSeek

37.5

Speed: —Context: 1M

Details

#83

MiMo-V2.5Open

Xiaomi

37.2

Speed: 81 t/sContext: 1M

Details

#84

Qwen3.6 27B (Reasoning)Open

Alibaba

37.1

Speed: 59 t/sContext: 262k

Details

#85

Grok 4.20 0309 v2 (Reasoning)

SpaceXAI

37.0

Speed: 181 t/sContext: 2M

Details

#86

GPT-5.1 (high)

OpenAI

36.9

Speed: 106 t/sContext: 272k

Details

#87

Grok 4.20 0309 (Reasoning)

SpaceXAI

36.5

Speed: —Context: 2M

Details

#88

Gemini 3.5 Flash-Lite

Google

36.5

Speed: 391 t/sContext: 1M

Details

#89

Claude 4.5 Sonnet (Reasoning)

Anthropic

36.4

Speed: 47 t/sContext: 1M

Details

#90

MiMo-V2-Omni-0327

Xiaomi

36.4

Speed: —Context: 256k

Details

#91

GPT-5 Codex (high)

OpenAI

36.1

Speed: —Context: 400k

Details

#92

Grok 4.3 (medium)

SpaceXAI

36.0

Speed: 140 t/sContext: 1M

Details

#93

Claude Sonnet 4.6 (Non-reasoning, High Effort)

Anthropic

35.9

Speed: 43 t/sContext: 1M

Details

#94

Grok 4.3 (low)

SpaceXAI

35.4

Speed: 128 t/sContext: 1M

Details

#95

GPT-5.5 (Non-reasoning)

OpenAI

35.4

Speed: 68 t/sContext: 922k

Details

#96

Kimi K2.5 (Reasoning)Open

Kimi

35.4

Speed: 52 t/sContext: 256k

Details

#97

GLM-5.1 (Non-reasoning)Open

Z AI

35.4

Speed: 64 t/sContext: 200k

Details

#98

MiMo-V2-Omni

Xiaomi

35.0

Speed: —Context: 256k

Details

#99

Gemini 3.5 Flash (minimal)

Google

34.9

Speed: 170 t/sContext: 1M

Details

#100

GPT-5.1 Codex (high)

OpenAI

34.7

Speed: —Context: 400k

Details

#101

GPT-5 (high)

OpenAI

34.7

Speed: 109 t/sContext: 400k

Details

#102

Claude Opus 4.5 (Non-reasoning)

Anthropic

34.7

Speed: 46 t/sContext: 200k

Details

#103

Kimi K2.6 (Non-reasoning)Open

Kimi

34.6

Speed: 32 t/sContext: 256k

Details

#104

GLM 5V Turbo (Reasoning)

Z AI

34.5

Speed: —Context: 200k

Details

#105

Claude Sonnet 4.6 (Non-reasoning, Low Effort)

Anthropic

34.3

Speed: 44 t/sContext: 1M

Details

#106

GLM-5.2 (Non-reasoning)Open

Z AI

34.1

Speed: 108 t/sContext: 1M

Details

#107

GPT-5.6 Terra (Non-reasoning)

OpenAI

34.0

Speed: 115 t/sContext: 1M

Details

#108

Qwen3.5 27B (Reasoning)Open

Alibaba

33.8

Speed: 77 t/sContext: 262k

Details

#109

GPT-5 (medium)

OpenAI

33.7

Speed: 90 t/sContext: 400k

Details

#110

Claude 4.1 Opus (Reasoning)

Anthropic

33.7

Speed: —Context: 200k

Details

#111

GLM-4.7 (Reasoning)Open

Z AI

33.7

Speed: 104 t/sContext: 200k

Details

#112

KAT Coder Pro V2

KwaiKAT

33.7

Speed: 107 t/sContext: 256k

Details

#113

Qwen3.5 397B A17B (Reasoning)Open

Alibaba

33.7

Speed: 68 t/sContext: 262k

Details

#114

MiniMax-M2.5Open

MiniMax

33.7

Speed: 98 t/sContext: 205k

Details

#115

Hy3-preview (Reasoning)Open

Tencent

33.6

Speed: —Context: 256k

Details

#116

GPT-5.5 Instant (May 2026)

OpenAI

33.5

Speed: —Context: 400k

Details

#117

LongCat 2.0Open

LongCat

33.5

Speed: 42 t/sContext: 1M

Details

#118

Grok 4

SpaceXAI

33.3

Speed: —Context: 256k

Details

#119

GPT-5.6 Luna (low)

OpenAI

33.3

Speed: 152 t/sContext: 1M

Details

#120

MiMo-V2-Flash (Feb 2026)Open

Xiaomi

33.2

Speed: —Context: 256k

Details

#121

Gemini 3 Pro Preview (low)

Google

33.1

Speed: —Context: 1M

Details

#122

Kimi K2 ThinkingOpen

Kimi

32.7

Speed: 123 t/sContext: 256k

Details

#123

o3-pro

OpenAI

32.5

Speed: —Context: 200k

Details

#124

GLM-5 (Non-reasoning)Open

Z AI

32.4

Speed: 40 t/sContext: 200k

Details

#125

Qwen3.5 122B A10B (Reasoning)Open

Alibaba

32.3

Speed: 133 t/sContext: 262k

Details

#126

DeepSeek V3.2 (Reasoning)Open

DeepSeek

32.0

Speed: —Context: 128k

Details

#127

Qwen3.5 397B A17B (Non-reasoning)Open

Alibaba

32.0

Speed: 65 t/sContext: 262k

Details

#128

Qwen3 Max Thinking

Alibaba

31.7

Speed: —Context: 256k

Details

#129

Qwen3.6 35B A3B (Reasoning)Open

Alibaba

31.6

Speed: 152 t/sContext: 262k

Details

#130

MiniMax-M2.1Open

MiniMax

31.4

Speed: 136 t/sContext: 205k

Details

#131

DeepSeek V4 Pro (Non-reasoning)Open

DeepSeek

31.2

Speed: 64 t/sContext: 1M

Details

#132

MiMo-V2-Flash (Reasoning)Open

Xiaomi

31.2

Speed: —Context: 256k

Details

#133

GPT-5 (low)

OpenAI

31.2

Speed: 84 t/sContext: 400k

Details

#134

Claude 4 Opus (Reasoning)

Anthropic

31.0

Speed: —Context: 200k

Details

#135

G9v3-39A5BOpen

AI9Stars

30.9

Speed: —Context: 131k

Details

#136

GPT-5 mini (medium)

OpenAI

30.9

Speed: 97 t/sContext: 400k

Details

#137

Qwen3.5 Omni Plus

Alibaba

30.6

Speed: 53 t/sContext: 256k

Details

#138

GPT-5.1 Codex mini (high)

OpenAI

30.6

Speed: —Context: 400k

Details

#139

Grok 4.1 Fast (Reasoning)

SpaceXAI

30.6

Speed: —Context: 2M

Details

#140

Ring-2.6-1TOpen

InclusionAI

30.6

Speed: —Context: 262k

Details

#141

Qwen3.6 27B (Non-reasoning)Open

Alibaba

30.5

Speed: 59 t/sContext: 262k

Details

#142

DeepSeek V3.1 Terminus (Reasoning)Open

DeepSeek

30.4

Speed: —Context: 128k

Details

#143

OpenAI

30.4

Speed: 127 t/sContext: 200k

Details

#144

Step 3.7 FlashOpen

StepFun

30.3

Speed: 395 t/sContext: 262k

Details

#145

GPT-5.4 nano (medium)

OpenAI

30.2

Speed: 165 t/sContext: 400k

Details

#146

Mistral Medium 3.5Open

Mistral

29.9

Speed: 143 t/sContext: 256k

Details

#147

GPT-5.4 mini (medium)

OpenAI

29.8

Speed: 189 t/sContext: 400k

Details

#148

Claude 4.5 Haiku (Reasoning)

Anthropic

29.6

Speed: 98 t/sContext: 200k

Details

#149

Kimi K2.5 (Non-reasoning)Open

Kimi

29.4

Speed: 52 t/sContext: 256k

Details

#150

Gemma 4 31B (Reasoning)Open

Google

29.4

Speed: 35 t/sContext: 256k

Details

#151

Qwen3.5 27B (Non-reasoning)Open

Alibaba

29.3

Speed: 83 t/sContext: 262k

Details

#152

Claude 4.5 Sonnet (Non-reasoning)

Anthropic

29.3

Speed: 41 t/sContext: 1M

Details

#153

Qwen3.5 35B A3B (Reasoning)Open

Alibaba

29.3

Speed: 156 t/sContext: 262k

Details

#154

Claude 4 Sonnet (Reasoning)

Anthropic

28.9

Speed: —Context: 1M

Details

#155

GPT-5.5 Instant (June 2026)

OpenAI

28.9

Speed: 170 t/sContext: 400k

Details

#156

GLM-4.6 (Reasoning)Open

Z AI

28.7

Speed: 49 t/sContext: 200k

Details

#157

DeepSeek V4 Flash (Non-reasoning)Open

DeepSeek

28.7

Speed: 121 t/sContext: 1M

Details

#158

JT-35B-Flash

China Mobile

28.4

Speed: —Context: 256k

Details

#159

MiniMax-M2Open

MiniMax

28.3

Speed: 101 t/sContext: 205k

Details

#160

KAT-Coder-Pro V1

KwaiKAT

28.3

Speed: —Context: 256k

Details

#161

Claude 4.1 Opus (Non-reasoning)

Anthropic

28.2

Speed: —Context: 200k

Details

#162

MiMo-V2.5-Pro (Non-reasoning)Open

Xiaomi

27.9

Speed: 45 t/sContext: 1M

Details

#163

GPT-5.4 (Non-reasoning)

OpenAI

27.7

Speed: 106 t/sContext: 1.1M

Details

#164

Qwen3.5 122B A10B (Non-reasoning)Open

Alibaba

27.6

Speed: 148 t/sContext: 262k

Details

#165

Grok 4 Fast (Reasoning)

SpaceXAI

27.4

Speed: —Context: 2M

Details

#166

Gemini 3 Flash Preview (Non-reasoning)

Google

27.4

Speed: 185 t/sContext: 1M

Details

#167

Claude 3.7 Sonnet (Reasoning)

Anthropic

27.1

Speed: —Context: 200k

Details

#168

GPT-5.6 Luna (Non-reasoning)

OpenAI

26.6

Speed: 164 t/sContext: 1M

Details

#169

GLM-4.7 (Non-reasoning)Open

Z AI

26.6

Speed: 101 t/sContext: 200k

Details

#170

Hy3-preview (Non-reasoning)Open

Tencent

26.1

Speed: —Context: 256k

Details

#171

Ling-2.6-1TOpen

InclusionAI

26.1

Speed: —Context: 262k

Details

#172

GPT-5.2 (Non-reasoning)

OpenAI

26.0

Speed: 73 t/sContext: 400k

Details

#173

Step 3.5 Flash 2603

StepFun

26.0

Speed: 155 t/sContext: 256k

Details

#174

Doubao Seed Code

ByteDance Seed

26.0

Speed: —Context: 256k

Details

#175

Gemini 2.5 Pro

Google

25.8

Speed: 143 t/sContext: 1M

Details

#176

Gemma 4 26B A4B (Reasoning)Open

Google

25.7

Speed: —Context: 256k

Details

#177

o4-mini (high)

OpenAI

25.6

Speed: 168 t/sContext: 200k

Details

#178

Step 3.5 FlashOpen

StepFun

25.5

Speed: 146 t/sContext: 256k

Details

#179

Claude 4 Opus (Non-reasoning)

Anthropic

25.5

Speed: —Context: 200k

Details

#180

Claude 4 Sonnet (Non-reasoning)

Anthropic

25.5

Speed: —Context: 1M

Details

#181

DeepSeek V3.2 Exp (Reasoning)Open

DeepSeek

25.4

Speed: —Context: 128k

Details

#182

NVIDIA Nemotron 3 Super 120B A12B (Reasoning)Open

NVIDIA

25.4

Speed: 143 t/sContext: 1M

Details

#183

GPT-5 mini (high)

OpenAI

25.3

Speed: 93 t/sContext: 400k

Details

#184

Gemini 3.1 Flash-Lite

Google

25.0

Speed: 312 t/sContext: 1M

Details

#185

Qwen3 Max Thinking (Preview)

Alibaba

25.0

Speed: 60 t/sContext: 262k

Details

#186

Grok 4.3 (Non-reasoning)

SpaceXAI

24.8

Speed: 134 t/sContext: 1M

Details

#187

MiMo-V2-Flash (Non-reasoning)Open

Xiaomi

24.7

Speed: —Context: 256k

Details

#188

DeepSeek V3.2 (Non-reasoning)Open

DeepSeek

24.7

Speed: —Context: 128k

Details

#189

Qwen3.6 35B A3B (Non-reasoning)Open

Alibaba

24.2

Speed: 166 t/sContext: 262k

Details

#190

Qwen3 Max

Alibaba

24.0

Speed: 64 t/sContext: 262k

Details

#191

Qwen3.5 35B A3B (Non-reasoning)Open

Alibaba

24.0

Speed: 175 t/sContext: 262k

Details

#192

gpt-oss-120b (high)Open

OpenAI

23.8

Speed: 199 t/sContext: 131k

Details

#193

Gemini 2.5 Flash Preview (Sep '25) (Reasoning)

Google

23.8

Speed: —Context: 1M

Details

#194

Claude 4.5 Haiku (Non-reasoning)

Anthropic

23.7

Speed: 90 t/sContext: 200k

Details

#195

Kimi K2 0905Open

Kimi

23.5

Speed: 37 t/sContext: 256k

Details

#196

Claude 3.7 Sonnet (Non-reasoning)

Anthropic

23.5

Speed: —Context: 200k

Details

#197

OpenAI

23.4

Speed: —Context: 200k

Details

#198

Gemini 2.5 Pro Preview (Mar' 25)

Google

23.0

Speed: —Context: 1M

Details

#199

GLM-4.6 (Non-reasoning)Open

Z AI

23.0

Speed: 59 t/sContext: 200k

Details

#200

GLM-4.7-Flash (Reasoning)Open

Z AI

22.9

Speed: 98 t/sContext: 200k

Details

#201

Command A+Open

Cohere

22.5

Speed: 198 t/sContext: 192k

Details

#202

Grok 3 mini Reasoning (high)

SpaceXAI

22.5

Speed: 81 t/sContext: 1M

Details

#203

Grok 4.20 0309 (Non-reasoning)

SpaceXAI

22.5

Speed: —Context: 2M

Details

#204

Gemini 2.5 Pro Preview (May' 25)

Google

22.3

Speed: —Context: 1M

Details

#205

DeepSeek V3.2 SpecialeOpen

DeepSeek

22.2

Speed: —Context: 128k

Details

#206

K-EXAONE (Reasoning)Open

LG AI Research

22.1

Speed: —Context: 256k

Details

#207

ERNIE 5.0 Thinking Preview

Baidu

21.9

Speed: —Context: 128k

Details

#208

Grok 4.20 0309 v2 (Non-reasoning)

SpaceXAI

21.8

Speed: 168 t/sContext: 2M

Details

#209

Gemma 4 12B (Reasoning)Open

Google

21.8

Speed: 111 t/sContext: 256k

Details

#210

Gemma 4 31B (Non-reasoning)Open

Google

21.8

Speed: 68 t/sContext: 256k

Details

#211

Nova 2.0 Pro Preview (medium)

Amazon

21.8

Speed: 120 t/sContext: 256k

Details

#212

Grok Code Fast 1

SpaceXAI

21.6

Speed: —Context: 256k

Details

#213

Qwen3.5 9B (Reasoning)Open

Alibaba

21.4

Speed: 83 t/sContext: 262k

Details

#214

Mercury 2

Inception

21.4

Speed: 788 t/sContext: 128k

Details

#215

DeepSeek V3.1 Terminus (Non-reasoning)Open

DeepSeek

21.4

Speed: —Context: 128k

Details

#216

DeepSeek V3.2 Exp (Non-reasoning)Open

DeepSeek

21.3

Speed: —Context: 128k

Details

#217

Apriel-v1.5-15B-ThinkerOpen

ServiceNow

21.2

Speed: —Context: 128k

Details

#218

Qwen3 Coder NextOpen

Alibaba

21.1

Speed: 133 t/sContext: 256k

Details

#219

DeepSeek V3.1 (Non-reasoning)Open

DeepSeek

21.0

Speed: —Context: 128k

Details

#220

Nova 2.0 Omni (medium)

Amazon

20.9

Speed: —Context: 1M

Details

#221

DeepSeek V3.1 (Reasoning)Open

DeepSeek

20.7

Speed: —Context: 128k

Details

#222

Qwen3 VL 235B A22B (Reasoning)Open

Alibaba

20.6

Speed: 58 t/sContext: 262k

Details

#223

Apriel-v1.6-15B-ThinkerOpen

ServiceNow

20.5

Speed: —Context: 128k

Details

#224

GPT-5.1 (Non-reasoning)

OpenAI

20.4

Speed: 102 t/sContext: 400k

Details

#225

Qwen3.5 9B (Non-reasoning)Open

Alibaba

20.3

Speed: —Context: 262k

Details

#226

EXAONE 4.5 33BOpen

LG AI Research

20.2

Speed: —Context: 262k

Details

#227

Gemma 4 26B A4B (Non-reasoning)Open

Google

20.1

Speed: 65 t/sContext: 256k

Details

#228

Qwen3.5 4B (Reasoning)Open

Alibaba

20.1

Speed: 45 t/sContext: 262k

Details

#229

DeepSeek R1 0528 (May '25)Open

DeepSeek

20.1

Speed: —Context: 128k

Details

#230

Gemini 2.5 Flash (Reasoning)

Google

20.1

Speed: 231 t/sContext: 1M

Details

#231

GPT-5 nano (high)

OpenAI

19.9

Speed: 150 t/sContext: 400k

Details

#232

North Mini CodeOpen

Cohere

19.8

Speed: 22 t/sContext: 256k

Details

#233

Qwen3 235B A22B 2507 (Reasoning)Open

Alibaba

19.6

Speed: 55 t/sContext: 256k

Details

#234

Nova 2.0 Pro Preview (low)

Amazon

19.6

Speed: 126 t/sContext: 256k

Details

#235

Mistral Small 4 (Reasoning)Open

Mistral

19.6

Speed: 160 t/sContext: 256k

Details

#236

GLM-4.5 (Reasoning)Open

Z AI

19.5

Speed: —Context: 128k

Details

#237

Kimi K2Open

Kimi

19.4

Speed: 38 t/sContext: 128k

Details

#238

GPT-4.1

OpenAI

19.4

Speed: 151 t/sContext: 1M

Details

#239

Qwen3 Max (Preview)

Alibaba

19.2

Speed: 59 t/sContext: 262k

Details

#240

Devstral 2Open

Mistral

19.2

Speed: 44 t/sContext: 256k

Details

#241

Nova 2.0 Lite (medium)

Amazon

19.0

Speed: 162 t/sContext: 1M

Details

#242

GPT-5 nano (medium)

OpenAI

19.0

Speed: 148 t/sContext: 400k

Details

#243

Qwen3.5 Omni Flash

Alibaba

19.0

Speed: 249 t/sContext: 256k

Details

#244

o3-mini

OpenAI

19.0

Speed: 215 t/sContext: 200k

Details

#245

o1-pro

OpenAI

18.9

Speed: —Context: 200k

Details

#246

Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)

Google

18.8

Speed: —Context: 1M

Details

#247

JT-MINI

China Mobile

18.5

Speed: —Context: 128k

Details

#248

DeepSeek R1 (Jan '25)Open

DeepSeek

18.5

Speed: —Context: 128k

Details

#249

Grok 3

SpaceXAI

18.4

Speed: —Context: 1M

Details

#250

Seed-OSS-36B-InstructOpen

ByteDance Seed

18.3

Speed: 35 t/sContext: 512k

Details

#251

Nova 2.0 Lite (high)

Amazon

18.2

Speed: 141 t/sContext: 1M

Details

#252

Trinity Large ThinkingOpen

Arcee AI

18.2

Speed: 191 t/sContext: 512k

Details

#253

Qwen3 235B A22B 2507 InstructOpen

Alibaba

18.2

Speed: 57 t/sContext: 256k

Details

#254

Qwen3 Coder 480B A35B InstructOpen

Alibaba

18.0

Speed: 66 t/sContext: 262k

Details

#255

Qwen3 VL 32B (Reasoning)Open

Alibaba

17.9

Speed: 90 t/sContext: 256k

Details

#256

Magistral Medium 1.2

Mistral

17.9

Speed: 114 t/sContext: 128k

Details

#257

Sonar Reasoning Pro

Perplexity

17.8

Speed: —Context: 127k

Details

#258

Nova 2.0 Lite (low)

Amazon

17.8

Speed: 157 t/sContext: 1M

Details

#259

HyperNova 60B 2605Open

Multiverse Computing

17.8

Speed: 347 t/sContext: 131k

Details

#260

MiniMax M1 80kOpen

MiniMax

17.7

Speed: —Context: 1M

Details

#261

GPT-5.4 nano (Non-Reasoning)

OpenAI

17.6

Speed: 165 t/sContext: 400k

Details

#262

Nemotron Cascade 2 30B A3BOpen

NVIDIA

17.6

Speed: —Context: 1M

Details

#263

Gemini 2.5 Flash Preview (Reasoning)

Google

17.5

Speed: —Context: 1M

Details

#264

Devstral Small 2Open

Mistral

17.4

Speed: 74 t/sContext: 256k

Details

#265

K2 Think V2Open

MBZUAI Institute of Foundation Models

17.3

Speed: —Context: 262k

Details

#266

LongCat Flash LiteOpen

LongCat

17.2

Speed: —Context: 256k

Details

#267

GPT-5 (minimal)

OpenAI

17.2

Speed: 78 t/sContext: 400k

Details

#268

o1-preview

OpenAI

17.0

Speed: —Context: 128k

Details

#269

HyperCLOVA X SEED Think (32B)Open

Naver

17.0

Speed: —Context: 128k

Details

#270

Grok 4.1 Fast (Non-reasoning)

SpaceXAI

16.9

Speed: —Context: 2M

Details

#271

GLM-4.6V (Reasoning)Open

Z AI

16.8

Speed: 77 t/sContext: 128k

Details

#272

K-EXAONE (Non-reasoning)Open

LG AI Research

16.7

Speed: —Context: 256k

Details

#273

Qwen3 Next 80B A3B (Reasoning)Open

Alibaba

16.7

Speed: 208 t/sContext: 262k

Details

#274

GPT-5.4 mini (Non-Reasoning)

OpenAI

16.6

Speed: 179 t/sContext: 400k

Details

#275

Nova 2.0 Omni (low)

Amazon

16.6

Speed: —Context: 1M

Details

#276

GLM-4.5-AirOpen

Z AI

16.5

Speed: 86 t/sContext: 128k

Details

#277

Grok 4 Fast (Non-reasoning)

SpaceXAI

16.5

Speed: —Context: 2M

Details

#278

Mi:dm K 2.5 Pro

Korea Telecom

16.4

Speed: —Context: 128k

Details

#279

Ring-1TOpen

InclusionAI

16.2

Speed: —Context: 128k

Details

#280

G9v3-3BOpen

AI9Stars

16.1

Speed: —Context: 131k

Details

#281

Qwen3.5 4B (Non-reasoning)Open

Alibaba

16.0

Speed: 25 t/sContext: 262k

Details

#282

Mistral Large 3Open

Mistral

15.9

Speed: 47 t/sContext: 256k

Details

#283

o3-mini (high)

OpenAI

15.6

Speed: 229 t/sContext: 200k

Details

#284

INTELLECT-3Open

Prime Intellect

15.6

Speed: —Context: 131k

Details

#285

GLM-4.7-Flash (Non-reasoning)Open

Z AI

15.5

Speed: 118 t/sContext: 200k

Details

#286

DeepSeek V3 0324Open

DeepSeek

15.4

Speed: —Context: 128k

Details

#287

GPT-5 (ChatGPT)

OpenAI

15.3

Speed: —Context: 128k

Details

#288

Solar Open 100B (Reasoning)Open

Upstage

15.1

Speed: —Context: 128k

Details

#289

Grok 3 Reasoning Beta

SpaceXAI

15.1

Speed: —Context: 1M

Details

#290

Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)

Google

15.1

Speed: —Context: 1M

Details

#291

Nemotron 3 Nano Omni 30B A3B ReasoningOpen

NVIDIA

14.9

Speed: 323 t/sContext: 256k

Details

#292

gpt-oss-120b (low)Open

OpenAI

14.9

Speed: 266 t/sContext: 131k

Details

#293

gpt-oss-20b (high)Open

OpenAI

14.9

Speed: 210 t/sContext: 131k

Details

#294

GPT-4.1 mini

OpenAI

14.8

Speed: 76 t/sContext: 1M

Details

#295

Mistral Small 3.1Open

Mistral

14.7

Speed: 149 t/sContext: 128k

Details

#296

Mistral Medium 3.1

Mistral

14.7

Speed: 84 t/sContext: 128k

Details

#297

Qwen3 30B A3B 2507 (Reasoning)Open

Alibaba

14.4

Speed: 137 t/sContext: 262k

Details

#298

MiniMax M1 40kOpen

MiniMax

14.4

Speed: —Context: 1M

Details

#299

Nova 2.0 Pro Preview (Non-reasoning)

Amazon

14.4

Speed: 103 t/sContext: 256k

Details

#300

gpt-oss-20b (low)Open

OpenAI

14.3

Speed: 216 t/sContext: 131k

Details

#301

Qwen3 VL 235B A22B InstructOpen

Alibaba

14.3

Speed: 53 t/sContext: 262k

Details

#302

Llama 4 MaverickOpen

Meta

14.3

Speed: 113 t/sContext: 1M

Details

#303

GPT-5 mini (minimal)

OpenAI

14.3

Speed: 100 t/sContext: 400k

Details

#304

K2-V2 (high)Open

MBZUAI Institute of Foundation Models

14.2

Speed: —Context: 512k

Details

#305

NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)Open

NVIDIA

14.2

Speed: 263 t/sContext: 1M

Details

#306

DeepSeek V3 (Dec '24)Open

DeepSeek

14.2

Speed: —Context: 128k

Details

#307

Gemini 2.5 Flash (Non-reasoning)

Google

14.1

Speed: 218 t/sContext: 1M

Details

#308

Solar Pro 3

Upstage

14.1

Speed: 121 t/sContext: 128k

Details

#309

Ling 2.6 FlashOpen

InclusionAI

14.1

Speed: 91 t/sContext: 262k

Details

#310

o1-mini

OpenAI

14.0

Speed: —Context: 128k

Details

#311

Qwen3 Next 80B A3B InstructOpen

Alibaba

13.7

Speed: 185 t/sContext: 262k

Details

#312

Tri-21B-think PreviewOpen

Trillion Labs

13.6

Speed: —Context: 32k

Details

#313

Qwen3 Coder 30B A3B InstructOpen

Alibaba

13.6

Speed: 94 t/sContext: 262k

Details

#314

GPT-4.5 (Preview)

OpenAI

13.6

Speed: —Context: 128k

Details

#315

DiffusionGemma 26B A4BOpen

Google

13.5

Speed: —Context: 256k

Details

#316

Qwen3 235B A22B (Reasoning)Open

Alibaba

13.4

Speed: 61 t/sContext: 33k

Details

#317

QwQ 32BOpen

Alibaba

13.4

Speed: —Context: 131k

Details

#318

Qwen3 VL 30B A3B (Reasoning)Open

Alibaba

13.3

Speed: 110 t/sContext: 256k

Details

#319

Gemini 2.0 Flash Thinking Experimental (Jan '25)

Google

13.3

Speed: —Context: 1M

Details

#320

Gemma 4 12B (Non-reasoning)Open

Google

13.2

Speed: 108 t/sContext: 262k

Details

#321

Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)

Google

13.1

Speed: —Context: 1M

Details

#322

Motif-2-12.7B-Reasoning

Motif Technologies

12.8

Speed: —Context: 128k

Details

#323

Ling-1TOpen

InclusionAI

12.8

Speed: —Context: 128k

Details

#324

Nova Premier

Amazon

12.7

Speed: 32 t/sContext: 1M

Details

#325

Solar Pro 2 (Preview) (Reasoning)

Upstage

12.5

Speed: —Context: 64k

Details

#326

Magistral Medium 1

Mistral

12.5

Speed: —Context: 40k

Details

#327

Mistral Medium 3

Mistral

12.5

Speed: 48 t/sContext: 128k

Details

#328

K2-V2 (medium)Open

MBZUAI Institute of Foundation Models

12.4

Speed: —Context: 512k

Details

#329

Llama Nemotron Super 49B v1.5 (Reasoning)Open

NVIDIA

12.4

Speed: 84 t/sContext: 128k

Details

#330

Devstral Medium

Mistral

12.4

Speed: —Context: 256k

Details

#331

Mistral Small 4 (Non-reasoning)Open

Mistral

12.4

Speed: 147 t/sContext: 256k

Details

#332

Tri-21B-ThinkOpen

Trillion Labs

12.4

Speed: —Context: 32k

Details

#333

GPT-4o (March 2025, chatgpt-4o-latest)

OpenAI

12.3

Speed: —Context: 128k

Details

#334

Gemini 2.0 Flash (Feb '25)

Google

12.3

Speed: —Context: 1M

Details

#335

Claude 3.5 Haiku

Anthropic

12.3

Speed: —Context: 200k

Details

#336

Llama 3.3 Nemotron Super 49B v1 (Reasoning)Open

NVIDIA

12.2

Speed: —Context: 128k

Details

#337

Gemma 4 E4B (Reasoning)Open

Google

12.2

Speed: 85 t/sContext: 128k

Details

#338

Qwen3 4B 2507 (Reasoning)Open

Alibaba

12.0

Speed: —Context: 262k

Details

#339

MiniCPM5-1B (Reasoning)Open

OpenBMB

12.0

Speed: —Context: 128k

Details

#340

Sarvam 105B (high)Open

Sarvam

11.9

Speed: —Context: 128k

Details

#341

Gemini 2.0 Pro Experimental (Feb '25)

Google

11.8

Speed: —Context: 2M

Details

#342

Nova 2.0 Lite (Non-reasoning)

Amazon

11.8

Speed: 150 t/sContext: 1M

Details

#343

Devstral Small (May '25)Open

Mistral

11.8

Speed: —Context: 256k

Details

#344

Celeris-1

Celeris

11.8

Speed: 2158 t/sContext: 131k

Details

#345

Claude 3 Opus

Anthropic

11.8

Speed: —Context: 200k

Details

#346

MiniCPM5-1B (Non-reasoning)Open

OpenBMB

11.7

Speed: —Context: 128k

Details

#347

Sonar Reasoning

Perplexity

11.7

Speed: —Context: 127k

Details

#348

Gemini 2.5 Flash Preview (Non-reasoning)

Google

11.7

Speed: —Context: 1M

Details

#349

Qwen3 32B (Reasoning)Open

Alibaba

11.5

Speed: 94 t/sContext: 33k

Details

#350

Gemini 2.5 Flash-Lite (Reasoning)

Google

11.4

Speed: 358 t/sContext: 1M

Details

#351

Magistral Small 1.2Open

Mistral

11.3

Speed: 147 t/sContext: 128k

Details

#352

GPT-4o (Nov '24)

OpenAI

11.2

Speed: 162 t/sContext: 128k

Details

#353

Nanbeige4.1-3BOpen

Nanbeige

11.1

Speed: —Context: 256k

Details

#354

Qwen3 VL 32B InstructOpen

Alibaba

11.1

Speed: 76 t/sContext: 256k

Details

#355

Ministral 3 14BOpen

Mistral

11.1

Speed: 97 t/sContext: 256k

Details

#356

DeepSeek R1 Distill Qwen 32BOpen

DeepSeek

11.0

Speed: —Context: 128k

Details

#357

GLM-4.6V (Non-reasoning)Open

Z AI

11.0

Speed: 85 t/sContext: 128k

Details

#358

Qwen3 235B A22B (Non-reasoning)Open

Alibaba

10.9

Speed: 61 t/sContext: 33k

Details

#359

Magistral Small 1Open

Mistral

10.7

Speed: —Context: 40k

Details

#360

Gemini 2.0 Flash (experimental)

Google

10.7

Speed: —Context: 1M

Details

#361

EXAONE 4.0 32B (Reasoning)Open

LG AI Research

10.6

Speed: —Context: 131k

Details

#362

Qwen3 VL 8B (Reasoning)Open

Alibaba

10.6

Speed: 113 t/sContext: 256k

Details

#363

Mistral Small 3.2Open

Mistral

10.6

Speed: 138 t/sContext: 128k

Details

#364

Nova 2.0 Omni (Non-reasoning)

Amazon

10.5

Speed: —Context: 1M

Details

#365

Qwen3 14B (Reasoning)Open

Alibaba

10.4

Speed: 45 t/sContext: 33k

Details

#366

DeepSeek R1 0528 Qwen3 8BOpen

DeepSeek

10.4

Speed: —Context: 33k

Details

#367

Qwen2.5 Max

Alibaba

10.2

Speed: —Context: 32k

Details

#368

Llama 4 ScoutOpen

Meta

10.0

Speed: 131 t/sContext: 10M

Details

#369

Qwen3 VL 30B A3B InstructOpen

Alibaba

10.0

Speed: 114 t/sContext: 256k

Details

#370

Hermes 4 - Llama-3.1 70B (Reasoning)Open

Nous Research

10.0

Speed: 90 t/sContext: 128k

Details

#371

Gemini 1.5 Pro (Sep '24)

Google

10.0

Speed: —Context: 2M

Details

#372

Solar Pro 2 (Preview) (Non-reasoning)

Upstage

10.0

Speed: —Context: 64k

Details

#373

DeepSeek R1 Distill Llama 70BOpen

DeepSeek

9.9

Speed: 28 t/sContext: 128k

Details

#374

Claude 3.5 Sonnet (Oct '24)

Anthropic

9.9

Speed: —Context: 200k

Details

#375

DeepSeek R1 Distill Qwen 14BOpen

DeepSeek

9.8

Speed: —Context: 128k

Details

#376

Falcon-H1R-7BOpen

TII UAE

9.8

Speed: —Context: 256k

Details

#377

Ling-flash-2.0Open

InclusionAI

9.7

Speed: 68 t/sContext: 128k

Details

#378

Gemma 4 E2B (Reasoning)Open

Google

9.7

Speed: —Context: 128k

Details

#379

Qwen3 Omni 30B A3B (Reasoning)Open

Alibaba

9.6

Speed: 100 t/sContext: 66k

Details

#380

GPT-4o (Aug '24)

OpenAI

9.6

Speed: 92 t/sContext: 128k

Details

#381

GPT-4.1 nano

OpenAI

9.6

Speed: 192 t/sContext: 1M

Details

#382

Qwen2.5 Instruct 72BOpen

Alibaba

9.6

Speed: —Context: 131k

Details

#383

Sonar

Perplexity

9.5

Speed: —Context: 127k

Details

#384

Step3 VL 10BOpen

StepFun

9.5

Speed: —Context: 66k

Details

#385

Llama 3.3 Instruct 70BOpen

Meta

9.4

Speed: 84 t/sContext: 128k

Details

#386

Qwen3 30B A3B (Reasoning)Open

Alibaba

9.3

Speed: 104 t/sContext: 33k

Details

#387

Sonar Pro

Perplexity

9.3

Speed: —Context: 200k

Details

#388

Devstral Small (Jul '25)Open

Mistral

9.3

Speed: —Context: 256k

Details

#389

QwQ 32B-PreviewOpen

Alibaba

9.2

Speed: —Context: 33k

Details

#390

GLM-4.5V (Reasoning)Open

Z AI

9.1

Speed: 82 t/sContext: 64k

Details

#391

Mistral Large 2 (Nov '24)Open

Mistral

9.1

Speed: —Context: 128k

Details

#392

Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)Open

NVIDIA

9.1

Speed: 52 t/sContext: 128k

Details

#393

Qwen3 30B A3B 2507 InstructOpen

Alibaba

9.1

Speed: 142 t/sContext: 262k

Details

#394

ERNIE 4.5 300B A47BOpen

Baidu

9.0

Speed: —Context: 131k

Details

#395

Hermes 4 - Llama-3.1 405B (Reasoning)Open

Nous Research

9.0

Speed: 34 t/sContext: 128k

Details

#396

Solar Pro 2 (Reasoning)

Upstage

9.0

Speed: —Context: 66k

Details

#397

NVIDIA Nemotron Nano 12B v2 VL (Reasoning)Open

NVIDIA

9.0

Speed: 95 t/sContext: 128k

Details

#398

Ministral 3 8BOpen

Mistral

9.0

Speed: 118 t/sContext: 256k

Details

#399

Gemma 4 E4B (Non-reasoning)Open

Google

8.9

Speed: 87 t/sContext: 128k

Details

#400

Granite 4.1 30BOpen

IBM

8.9

Speed: —Context: 131k

Details

#401

NVIDIA Nemotron Nano 9B V2 (Reasoning)Open

NVIDIA

8.8

Speed: 192 t/sContext: 131k

Details

#402

Hermes 4 - Llama-3.1 405B (Non-reasoning)Open

Nous Research

8.8

Speed: 33 t/sContext: 128k

Details

#403

Gemini 2.0 Flash-Lite (Feb '25)

Google

8.8

Speed: —Context: 1M

Details

#404

NVIDIA Nemotron 3 Nano 4BOpen

NVIDIA

8.7

Speed: —Context: 262k

Details

#405

Llama Nemotron Super 49B v1.5 (Non-reasoning)Open

NVIDIA

8.7

Speed: 91 t/sContext: 128k

Details

#406

Qwen3 32B (Non-reasoning)Open

Alibaba

8.6

Speed: 97 t/sContext: 33k

Details

#407

GPT-4o (May '24)

OpenAI

8.6

Speed: 101 t/sContext: 128k

Details

#408

Gemini 2.0 Flash-Lite (Preview)

Google

8.6

Speed: —Context: 1M

Details

#409

K2-V2 (low)Open

MBZUAI Institute of Foundation Models

8.6

Speed: —Context: 512k

Details

#410

Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)Open

NVIDIA

8.5

Speed: —Context: 128k

Details

#411

Kimi Linear 48B A3B InstructOpen

Kimi

8.5

Speed: —Context: 1M

Details

#412

Llama 3.1 Instruct 405BOpen

Meta

8.5

Speed: —Context: 128k

Details

#413

Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)Open

NVIDIA

8.5

Speed: —Context: 128k

Details

#414

Qwen3 VL 8B InstructOpen

Alibaba

8.4

Speed: 119 t/sContext: 256k

Details

#415

Qwen3 4B (Reasoning)Open

Alibaba

8.4

Speed: —Context: 32k

Details

#416

LFM2.5-8B-A1BOpen

Liquid AI

8.3

Speed: 337 t/sContext: 33k

Details

#417

Claude 3.5 Sonnet (June '24)

Anthropic

8.3

Speed: —Context: 200k

Details

#418

Qwen3 8B (Reasoning)Open

Alibaba

8.3

Speed: 39 t/sContext: 131k

Details

#419

Llama 3.1 Tulu3 405BOpen

Allen Institute for AI

8.3

Speed: —Context: 128k

Details

#420

GPT-4o (ChatGPT)

OpenAI

8.2

Speed: —Context: 128k

Details

#421

Ring-flash-2.0Open

InclusionAI

8.2

Speed: —Context: 128k

Details

#422

Pixtral LargeOpen

Mistral

8.1

Speed: —Context: 128k

Details

#423

Olmo 3.1 32B ThinkOpen

Allen Institute for AI

8.1

Speed: —Context: 66k

Details

#424

Grok 2 (Dec '24)Open

SpaceXAI

8.0

Speed: —Context: 131k

Details

#425

GPT-5 nano (minimal)

OpenAI

8.0

Speed: 144 t/sContext: 400k

Details

#426

Gemini 1.5 Flash (Sep '24)

Google

8.0

Speed: —Context: 1M

Details

#427

Qwen3 VL 4B (Reasoning)Open

Alibaba

7.9

Speed: —Context: 256k

Details

#428

GPT-4 Turbo

OpenAI

7.9

Speed: 32 t/sContext: 128k

Details

#429

Solar Pro 2 (Non-reasoning)

Upstage

7.8

Speed: —Context: 66k

Details

#430

Nova Pro

Amazon

7.7

Speed: —Context: 300k

Details

#431

Command AOpen

Cohere

7.7

Speed: 59 t/sContext: 256k

Details

#432

Qwen3.5 2B (Reasoning)Open

Alibaba

7.6

Speed: —Context: 262k

Details

#433

Llama 3.1 Nemotron Instruct 70BOpen

NVIDIA

7.6

Speed: 96 t/sContext: 128k

Details

#434

Llama 3.1 Instruct 8BOpen

Meta

7.6

Speed: 141 t/sContext: 128k

Details

#435

Grok Beta

SpaceXAI

7.5

Speed: —Context: 128k

Details

#436

Qwen2.5 Instruct 32BOpen

Alibaba

7.5

Speed: —Context: 128k

Details

#437

Gemma 3 27B InstructOpen

Google

7.4

Speed: —Context: 128k

Details

#438

NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)Open

NVIDIA

7.4

Speed: 205 t/sContext: 1M

Details

#439

NVIDIA Nemotron Nano 9B V2 (Non-reasoning)Open

NVIDIA

7.4

Speed: 172 t/sContext: 131k

Details

#440

Mistral Large 2 (Jul '24)Open

Mistral

7.3

Speed: —Context: 128k

Details

#441

Qwen3 4B 2507 InstructOpen

Alibaba

7.1

Speed: —Context: 262k

Details

#442

Qwen2.5 Coder Instruct 32BOpen

Alibaba

7.1

Speed: —Context: 131k

Details

#443

Qwen3 14B (Non-reasoning)Open

Alibaba

7.0

Speed: 56 t/sContext: 33k

Details

#444

GPT-4

OpenAI

7.0

Speed: —Context: 8k

Details

#445

GLM-4.5V (Non-reasoning)Open

Z AI

7.0

Speed: 81 t/sContext: 64k

Details

#446

Mistral Small 3Open

Mistral

6.9

Speed: 146 t/sContext: 32k

Details

#447

Gemini 2.5 Flash-Lite (Non-reasoning)

Google

6.9

Speed: 232 t/sContext: 1M

Details

#448

Nova Lite

Amazon

6.9

Speed: 162 t/sContext: 300k

Details

#449

GPT-4o mini

OpenAI

6.9

Speed: 82 t/sContext: 128k

Details

#450

Hermes 4 - Llama-3.1 70B (Non-reasoning)Open

Nous Research

6.9

Speed: 95 t/sContext: 128k

Details

#451

Qwen3 30B A3B (Non-reasoning)Open

Alibaba

6.8

Speed: 108 t/sContext: 33k

Details

#452

Ministral 3 3BOpen

Mistral

6.8

Speed: 203 t/sContext: 256k

Details

#453

DeepSeek-V2.5 (Dec '24)Open

DeepSeek

6.8

Speed: —Context: 128k

Details

#454

Qwen3 4B (Non-reasoning)Open

Alibaba

6.8

Speed: —Context: 32k

Details

#455

Llama 3.1 Instruct 70BOpen

Meta

6.8

Speed: 61 t/sContext: 128k

Details

#456

Granite 4.1 8BOpen

IBM

6.7

Speed: 111 t/sContext: 131k

Details

#457

Sarvam 30B (high)Open

Sarvam

6.6

Speed: —Context: 66k

Details

#458

Gemini 2.0 Flash Thinking Experimental (Dec '24)

Google

6.6

Speed: —Context: 2M

Details

#459

DeepSeek-V2.5Open

DeepSeek

6.6

Speed: —Context: 128k

Details

#460

Olmo 3.1 32B InstructOpen

Allen Institute for AI

6.5

Speed: —Context: 66k

Details

#461

Mistral Saba

Mistral

6.4

Speed: —Context: 32k

Details

#462

DeepSeek R1 Distill Llama 8BOpen

DeepSeek

6.4

Speed: —Context: 128k

Details

#463

Gemma 4 E2B (Non-reasoning)Open

Google

6.4

Speed: —Context: 128k

Details

#464

Olmo 3 32B ThinkOpen

Allen Institute for AI

6.4

Speed: —Context: 66k

Details

#465

Gemini 1.5 Pro (May '24)

Google

6.3

Speed: —Context: 2M

Details

#466

R1 1776Open

Perplexity

6.3

Speed: —Context: 128k

Details

#467

Qwen2.5 Turbo

Alibaba

6.3

Speed: 107 t/sContext: 1M

Details

#468

Reka Flash (Sep '24)

Reka AI

6.3

Speed: 57 t/sContext: 128k

Details

#469

Llama 3.2 Instruct 90B (Vision)Open

Meta

6.2

Speed: —Context: 128k

Details

#470

Solar MiniOpen

Upstage

6.2

Speed: —Context: 4k

Details

#471

Grok-1Open

SpaceXAI

6.0

Speed: —Context: 8k

Details

#472

Qwen2 Instruct 72BOpen

Alibaba

6.0

Speed: —Context: 131k

Details

#473

Phi-4 Mini InstructOpen

Microsoft

6.0

Speed: 44 t/sContext: 128k

Details

#474

EXAONE 4.0 32B (Non-reasoning)Open

LG AI Research

6.0

Speed: —Context: 131k

Details

#475

Qwen3.5 2B (Non-reasoning)Open

Alibaba

5.6

Speed: —Context: 262k

Details

#476

Gemini 1.5 Flash-8B

Google

5.5

Speed: —Context: 1M

Details

#477

Gemma 3 12B InstructOpen

Google

5.5

Speed: —Context: 128k

Details

#478

Qwen3.5 0.8B (Reasoning)Open

Alibaba

5.5

Speed: —Context: 262k

Details

#479

DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Open

Nous Research

5.3

Speed: —Context: 32k

Details

#480

Jamba 1.7 LargeOpen

AI21 Labs

5.3

Speed: 58 t/sContext: 256k

Details

#481

Granite 4.0 H SmallOpen

IBM

5.2

Speed: 48 t/sContext: 128k

Details

#482

Jamba 1.5 LargeOpen

AI21 Labs

5.1

Speed: —Context: 256k

Details

#483

Qwen3 Omni 30B A3B InstructOpen

Alibaba

5.1

Speed: 95 t/sContext: 66k

Details

#484

Hermes 3 - Llama-3.1 70BOpen

Nous Research

5.1

Speed: 33 t/sContext: 128k

Details

#485

Qwen3 8B (Non-reasoning)Open

Alibaba

5.1

Speed: 39 t/sContext: 33k

Details

#486

DeepSeek-Coder-V2Open

DeepSeek

5.1

Speed: —Context: 128k

Details

#487

OLMo 2 32BOpen

Allen Institute for AI

5.0

Speed: —Context: 4k

Details

#488

Jamba 1.6 LargeOpen

AI21 Labs

5.0

Speed: 56 t/sContext: 256k

Details

#489

LFM2 24B A2BOpen

Liquid AI

4.9

Speed: —Context: 33k

Details

#490

Gemini 1.5 Flash (May '24)

Google

4.9

Speed: —Context: 1M

Details

#491

Phi-4Open

Microsoft

4.9

Speed: —Context: 16k

Details

#492

Claude 3 Sonnet

Anthropic

4.7

Speed: —Context: 200k

Details

#493

Nova Micro

Amazon

4.7

Speed: 279 t/sContext: 130k

Details

#494

Granite 4.1 3BOpen

IBM

4.7

Speed: —Context: 131k

Details

#495

Mistral Small (Sep '24)Open

Mistral

4.7

Speed: 147 t/sContext: 33k

Details

#496

Gemini 1.0 Ultra

Google

4.6

Speed: —Context: 33k

Details

#497

Phi-3 Mini Instruct 3.8BOpen

Microsoft

4.6

Speed: —Context: 4k

Details

#498

NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)Open

NVIDIA

4.6

Speed: 166 t/sContext: 128k

Details

#499

Gemma 3n E4B Instruct Preview (May '25)Open

Google

4.6

Speed: —Context: 32k

Details

#500

Phi-4 Multimodal InstructOpen

Microsoft

4.5

Speed: 18 t/sContext: 128k

Details

#501

Qwen2.5 Coder Instruct 7B Open

Alibaba

4.5

Speed: —Context: 131k

Details

#502

Mistral Large (Feb '24)

Mistral

4.4

Speed: —Context: 33k

Details

#503

Mixtral 8x22B InstructOpen

Mistral

4.4

Speed: —Context: 65k

Details

#504

Llama 2 Chat 7BOpen

Meta

4.3

Speed: —Context: 4k

Details

#505

Llama 3.2 Instruct 3BOpen

Meta

4.2

Speed: —Context: 128k

Details

#506

MiniCPM-V 4.6 1.3BOpen

OpenBMB

4.2

Speed: —Context: 262k

Details

#507

Jamba Reasoning 3BOpen

AI21 Labs

4.1

Speed: —Context: 262k

Details

#508

Qwen3 VL 4B InstructOpen

Alibaba

4.1

Speed: —Context: 256k

Details

#509

Qwen1.5 Chat 110BOpen

Alibaba

4.1

Speed: —Context: 32k

Details

#510

Reka Flash 3Open

Reka AI

4.1

Speed: —Context: 128k

Details

#511

Olmo 3 7B ThinkOpen

Allen Institute for AI

4.0

Speed: —Context: 66k

Details

#512

Claude 2.1

Anthropic

3.9

Speed: —Context: 200k

Details

#513

Claude 3 Haiku

Anthropic

3.9

Speed: —Context: 200k

Details

#514

OLMo 2 7BOpen

Allen Institute for AI

3.9

Speed: —Context: 4k

Details

#515

Molmo 7B-DOpen

Allen Institute for AI

3.8

Speed: —Context: 4k

Details

#516

Ling-mini-2.0Open

InclusionAI

3.8

Speed: —Context: 131k

Details

#517

DeepSeek R1 Distill Qwen 1.5BOpen

DeepSeek

3.7

Speed: —Context: 128k

Details

#518

DeepSeek-V2-ChatOpen

DeepSeek

3.6

Speed: —Context: 128k

Details

#519

Claude 2.0

Anthropic

3.6

Speed: —Context: 100k

Details

#520

Mistral Small (Feb '24)

Mistral

3.6

Speed: 145 t/sContext: 33k

Details

#521

Mistral Medium

Mistral

3.6

Speed: 133 t/sContext: 33k

Details

#522

GPT-3.5 Turbo

OpenAI

3.6

Speed: —Context: 4k

Details

#523

Llama 3 Instruct 70BOpen

Meta

3.5

Speed: —Context: 8k

Details

#524

Qwen Chat 72BOpen

Alibaba

3.4

Speed: —Context: 34k

Details

#525

Arctic InstructOpen

Snowflake

3.4

Speed: —Context: 4k

Details

#526

LFM 40B

Liquid AI

3.4

Speed: —Context: 32k

Details

#527

Llama 3.2 Instruct 11B (Vision)Open

Meta

3.3

Speed: 49 t/sContext: 128k

Details

#528

Qwen3.5 0.8B (Non-reasoning)Open

Alibaba

3.3

Speed: —Context: 262k

Details

#529

PALM-2

Google

3.2

Speed: —Context: 8k

Details

#530

Gemini 1.0 Pro

Google

3.1

Speed: —Context: 33k

Details

#531

DeepSeek Coder V2 Lite InstructOpen

DeepSeek

3.1

Speed: —Context: 128k

Details

#532

Sarvam M (Reasoning)Open

Sarvam

3.0

Speed: —Context: 33k

Details

#533

Llama 2 Chat 70BOpen

Meta

3.0

Speed: —Context: 4k

Details

#534

DeepSeek LLM 67B Chat (V1)Open

DeepSeek

3.0

Speed: —Context: 4k

Details

#535

Llama 2 Chat 13BOpen

Meta

3.0

Speed: —Context: 4k

Details

#536

Command-R+ (Apr '24)Open

Cohere

3.0

Speed: —Context: 128k

Details

#537

OpenChat 3.5 (1210)Open

OpenChat

3.0

Speed: —Context: 8k

Details

#538

DBRX InstructOpen

Databricks

3.0

Speed: —Context: 33k

Details

#539

Exaone 4.0 1.2B (Reasoning)Open

LG AI Research

2.9

Speed: —Context: 64k

Details

#540

Olmo 3 7B InstructOpen

Allen Institute for AI

2.8

Speed: —Context: 66k

Details

#541

Exaone 4.0 1.2B (Non-reasoning)Open

LG AI Research

2.8

Speed: —Context: 64k

Details

#542

LFM2.5-1.2B-ThinkingOpen

Liquid AI

2.7

Speed: —Context: 32k

Details

#543

Jamba 1.7 MiniOpen

AI21 Labs

2.7

Speed: —Context: 258k

Details

#544

LFM2 2.6BOpen

Liquid AI

2.7

Speed: —Context: 33k

Details

#545

LFM2.5-1.2B-InstructOpen

Liquid AI

2.7

Speed: —Context: 32k

Details

#546

Jamba 1.5 MiniOpen

AI21 Labs

2.7

Speed: —Context: 256k

Details

#547

Granite 4.0 H 1BOpen

IBM

2.7

Speed: —Context: 128k

Details

#548

Qwen3 1.7B (Reasoning)Open

Alibaba

2.6

Speed: —Context: 32k

Details

#549

Jamba 1.6 MiniOpen

AI21 Labs

2.6

Speed: 178 t/sContext: 256k

Details

#550

Mixtral 8x7B InstructOpen

Mistral

2.4

Speed: —Context: 33k

Details

#551

Gemma 3 270MOpen

Google

2.4

Speed: —Context: 32k

Details

#552

Apertus 70B InstructOpen

Swiss AI Initiative

2.4

Speed: —Context: 66k

Details

#553

Granite 4.0 MicroOpen

IBM

2.4

Speed: —Context: 128k

Details

#554

DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Open

Nous Research

2.3

Speed: —Context: 128k

Details

#555

Mistral 7B InstructOpen

Mistral

2.1

Speed: 120 t/sContext: 8k

Details

#556

Claude Instant

Anthropic

2.1

Speed: —Context: 100k

Details

#557

Command-R (Mar '24)Open

Cohere

2.1

Speed: —Context: 128k

Details

#558

Qwen Chat 14BOpen

Alibaba

2.1

Speed: —Context: 8k

Details

#559

Llama 65BOpen

Meta

2.1

Speed: —Context: 2k

Details

#560

Granite 4.0 1BOpen

IBM

2.1

Speed: —Context: 128k

Details

#561

Molmo2-8BOpen

Allen Institute for AI

2.0

Speed: —Context: 37k

Details

#562

LFM2 8B A1BOpen

Liquid AI

1.8

Speed: —Context: 33k

Details

#563

Granite 3.3 8B (Non-reasoning)Open

IBM

1.8

Speed: 15 t/sContext: 128k

Details

#564

Qwen3 1.7B (Non-reasoning)Open

Alibaba

1.5

Speed: —Context: 32k

Details

#565

Qwen3 0.6B (Reasoning)Open

Alibaba

1.3

Speed: —Context: 32k

Details

#566

Llama 3 Instruct 8BOpen

Meta

1.2

Speed: —Context: 8k

Details

#567

Gemma 3n E4B InstructOpen

Google

1.2

Speed: 43 t/sContext: 32k

Details

#568

LFM2 1.2BOpen

Liquid AI

1.1

Speed: —Context: 33k

Details

#569

Gemma 3 4B InstructOpen

Google

1.1

Speed: —Context: 128k

Details

#570

Llama 3.2 Instruct 1BOpen

Meta

1.1

Speed: —Context: 128k

Details

#571

LFM2.5-VL-1.6BOpen

Liquid AI

1.0

Speed: 398 t/sContext: 32k

Details

#572

Qwen3 0.6B (Non-reasoning)Open

Alibaba

1.0

Speed: —Context: 32k

Details

#573

Granite 4.0 350MOpen

IBM

1.0

Speed: —Context: 33k

Details

#574

Tiny Aya GlobalOpen

Cohere

1.0

Speed: —Context: 8k

Details

#575

Apertus 8B InstructOpen

Swiss AI Initiative

1.0

Speed: —Context: 66k

Details

#576

Granite 4.0 H 350MOpen

IBM

1.0

Speed: —Context: 33k

Details

#577

Gemma 3n E2B InstructOpen

Google

1.0

Speed: —Context: 32k

Details

#578

Gemma 3 1B InstructOpen

Google

1.0

Speed: —Context: 32k

Details

#	Model & Provider	Score	Speed	Latency	Context	Frontend	Game Dev	Agentic	Pricing /1M
1	Claude Opus 5 (Adaptive Reasoning, Max Effort) Anthropic	60.7	56 t/s	63ms	1M	60.7	60.1	59.5	In: $5.00 Out: $25.00
2	Claude Opus 5 (Adaptive Reasoning, Xhigh Effort) Anthropic	60.1	57 t/s	31ms	1M	60.1	59.5	58.9	In: $5.00 Out: $25.00
3	Claude Fable 5 (Adaptive Reasoning, Max Effort, Opus 4.8 Fallback) Anthropic	59.9	73 t/s	83ms	1M	59.9	59.9	59.9	In: $10.00 Out: $50.00
4	GPT-5.6 Sol (max) OpenAI	58.9	71 t/s	131ms	1M	58.9	58.9	58.9	In: $5.00 Out: $30.00
5	Claude Opus 5 (Adaptive Reasoning, High Effort) Anthropic	58.9	54 t/s	18ms	1M	58.9	58.9	58.9	In: $5.00 Out: $25.00
6	GPT-5.6 Sol (xhigh) OpenAI	57.7	72 t/s	48ms	1M	57.7	57.7	57.7	In: $5.00 Out: $30.00
7	Kimi K3 (max)Open Kimi	57.1	36 t/s	3ms	1M	57.1	57.1	57.1	In: $3.00 Out: $15.00
8	Claude Opus 5 (Adaptive Reasoning, Medium Effort) Anthropic	56.3	54 t/s	7ms	1M	56.3	56.3	56.3	In: $5.00 Out: $25.00
9	GPT-5.6 Sol (high) OpenAI	55.9	67 t/s	10ms	1M	55.9	55.9	55.9	In: $5.00 Out: $30.00
10	Claude Opus 4.8 (Adaptive Reasoning, Max Effort) Anthropic	55.7	58 t/s	23ms	1M	55.7	55.7	55.7	In: $5.00 Out: $25.00
11	GPT-5.6 Terra (max) OpenAI	55.0	143 t/s	158ms	1M	55.0	55.0	55.0	In: $2.00 Out: $12.00
12	GPT-5.5 (xhigh) OpenAI	54.8	87 t/s	66ms	922k	54.8	54.8	54.8	In: $5.00 Out: $30.00
13	Grok 4.5 (high) SpaceXAI	53.8	61 t/s	10ms	500k	53.8	53.8	53.8	In: $2.00 Out: $6.00
14	GPT-5.6 Sol (medium) OpenAI	53.6	64 t/s	5ms	1M	53.6	53.6	53.6	In: $5.00 Out: $30.00
15	Claude Opus 4.7 (Adaptive Reasoning, Max Effort) Anthropic	53.5	52 t/s	23ms	1M	53.5	53.5	53.5	In: $5.00 Out: $25.00
16	Claude Sonnet 5 (Adaptive Reasoning, Max Effort) Anthropic	53.4	81 t/s	201ms	1M	53.4	53.4	53.4	In: $2.00 Out: $10.00
17	GPT-5.5 (high) OpenAI	53.1	78 t/s	31ms	922k	53.1	53.1	53.1	In: $5.00 Out: $30.00
18	GPT-5.6 Terra (xhigh) OpenAI	51.6	124 t/s	15ms	1M	51.6	51.6	51.6	In: $2.00 Out: $12.00
19	GPT-5.4 (xhigh) OpenAI	51.4	143 t/s	130ms	1.1M	51.4	51.4	51.4	In: $2.50 Out: $15.00
20	GPT-5.6 Luna (max) OpenAI	51.2	181 t/s	133ms	1M	51.2	51.2	51.2	In: $0.20 Out: $1.20
21	Z GLM-5.2 (max)Open Z AI	51.1	186 t/s	2ms	1M	51.1	51.1	51.1	In: $1.35 Out: $4.29
22	Muse Spark 1.1 (xhigh) Meta	50.6	216 t/s	3ms	1M	50.6	50.6	50.6	In: $1.25 Out: $4.25
23	Claude Opus 5 (Adaptive Reasoning, Low Effort) Anthropic	50.6	56 t/s	3ms	1M	50.6	50.6	50.6	In: $5.00 Out: $25.00
24	GPT-5.5 (medium) OpenAI	50.4	74 t/s	8ms	922k	50.4	50.4	50.4	In: $5.00 Out: $30.00
25	Gemini 3.5 Flash (high) Google	50.2	187 t/s	22ms	1M	50.2	50.2	50.2	In: $1.50 Out: $9.00
26	Gemini 3.6 Flash (high) Google	50.1	213 t/s	16ms	1M	60.0	59.4	58.8	In: $1.50 Out: $7.50
27	DeepSeek V4 Flash 0731 (Reasoning, Max Effort)Open DeepSeek	49.9	124 t/s	1ms	1M	49.9	49.9	49.9	In: $0.14 Out: $0.28
28	GPT-5.6 Sol (low) OpenAI	49.4	65 t/s	3ms	1M	49.4	49.4	49.4	In: $5.00 Out: $30.00
29	GPT-5.6 Luna (xhigh) OpenAI	49.1	163 t/s	46ms	1M	49.1	49.1	49.1	In: $0.20 Out: $1.20
30	GPT-5.6 Terra (high) OpenAI	49.0	122 t/s	3ms	1M	49.0	49.0	49.0	In: $2.00 Out: $12.00
31	Claude Sonnet 4.6 (Adaptive Reasoning, Max Effort) Anthropic	47.2	56 t/s	79ms	1M	47.2	47.2	47.2	In: $3.00 Out: $15.00
32	Kimi K3 (low)Open Kimi	46.6	36 t/s	3ms	1M	46.6	46.6	46.6	In: $3.00 Out: $15.00
33	Gemini 3.1 Pro Preview Google	46.5	129 t/s	34ms	1M	46.5	46.5	46.5	In: $2.00 Out: $12.00
34	GPT-5.6 Luna (high) OpenAI	46.1	166 t/s	6ms	1M	46.1	46.1	46.1	In: $0.20 Out: $1.20
35	Qwen3.7 Max Alibaba	46.0	205 t/s	2ms	1M	46.0	46.0	46.0	In: $2.50 Out: $7.50
36	GPT-5.6 Terra (medium) OpenAI	45.6	120 t/s	2ms	1M	45.6	45.6	45.6	In: $2.00 Out: $12.00
37	Gemini 3.5 Flash (medium) Google	45.4	188 t/s	20ms	1M	45.4	45.4	45.4	In: $1.50 Out: $9.00
38	MiniMax-M3Open MiniMax	44.4	92 t/s	1ms	1M	44.4	44.4	44.4	In: $0.30 Out: $1.20
39	DeepSeek V4 Pro (Reasoning, Max Effort)Open DeepSeek	44.3	66 t/s	2ms	1M	44.3	44.3	44.3	In: $0.43 Out: $0.87
40	GPT-5.3 Codex (xhigh) OpenAI	44.3	130 t/s	64ms	400k	44.3	44.3	44.3	In: $1.75 Out: $14.00
41	Kimi K2.6Open Kimi	44.2	38 t/s	3ms	256k	44.2	44.2	44.2	In: $0.95 Out: $4.00
42	M Motif 3 (Beta) Motif Technologies	44.1	—	—	262k	44.1	44.1	44.1	In: — Out: —
43	Claude Opus 4.6 (Adaptive Reasoning, Max Effort) Anthropic	43.7	42 t/s	17ms	1M	43.7	43.7	43.7	In: $5.00 Out: $25.00
44	GPT-5.5 (low) OpenAI	43.5	72 t/s	2ms	922k	43.5	43.5	43.5	In: $5.00 Out: $30.00
45	DeepSeek V4 Pro (Reasoning, High Effort)Open DeepSeek	43.1	66 t/s	1ms	1M	43.1	43.1	43.1	In: $0.43 Out: $0.87
46	Muse Spark Meta	43.1	—	—	262k	43.1	43.1	43.1	In: — Out: —
47	Claude Opus 4.7 (Non-reasoning, High Effort) Anthropic	42.7	47 t/s	2ms	1M	42.7	42.7	42.7	In: $5.00 Out: $25.00
48	MiMo-V2.5-ProOpen Xiaomi	42.2	48 t/s	3ms	1M	42.2	42.2	42.2	In: $0.43 Out: $0.87
49	GPT-5.2 (xhigh) OpenAI	42.2	90 t/s	156ms	400k	42.2	42.2	42.2	In: $1.75 Out: $14.00
50	Kimi K2.7 CodeOpen Kimi	41.9	40 t/s	3ms	256k	41.9	41.9	41.9	In: $0.95 Out: $4.00
51	Claude Sonnet 5 (Non-reasoning, High Effort) Anthropic	41.7	65 t/s	2ms	1M	41.7	41.7	41.7	In: $2.00 Out: $10.00
52	T Hy3Open Tencent	41.2	70 t/s	3ms	256k	41.2	41.2	41.2	In: $0.14 Out: $0.56
53	GPT-5.6 Sol (Non-reasoning) OpenAI	41.2	64 t/s	1ms	1M	41.2	41.2	41.2	In: $5.00 Out: $30.00
54	N Nex-N2-ProOpen Nex AGI	41.0	130 t/s	1ms	262k	41.0	41.0	41.0	In: $0.50 Out: $2.50
55	Claude Opus 4.5 (Reasoning) Anthropic	40.8	49 t/s	14ms	200k	40.8	40.8	40.8	In: $5.00 Out: $25.00
56	T Inkling (xhigh)Open Thinking Machines	40.7	85 t/s	2ms	1M	40.7	40.7	40.7	In: $1.00 Out: $4.05
57	GPT-5.6 Terra (low) OpenAI	40.5	122 t/s	2ms	1M	40.5	40.5	40.5	In: $2.00 Out: $12.00
58	MiMo-V2-Pro Xiaomi	40.3	—	—	1M	40.3	40.3	40.3	In: — Out: —
59	DeepSeek V4 Flash (Reasoning, Max Effort)Open DeepSeek	40.3	—	—	1M	40.3	40.3	40.3	In: $0.14 Out: $0.28
60	T Inkling SmallOpen Thinking Machines	40.2	132 t/s	2ms	1M	40.2	40.2	40.2	In: $0.30 Out: $1.20
61	Z GLM-5.1 (Reasoning)Open Z AI	40.2	73 t/s	2ms	200k	40.2	40.2	40.2	In: $1.38 Out: $4.40
62	GPT-5.2 Codex (xhigh) OpenAI	40.1	—	—	400k	40.1	40.1	40.1	In: $1.75 Out: $14.00
63	Qwen3.6 Max Preview Alibaba	40.0	51 t/s	3ms	256k	40.0	40.0	40.0	In: $1.30 Out: $7.80
64	GPT-5.4 mini (xhigh) OpenAI	40.0	190 t/s	9ms	400k	40.0	40.0	40.0	In: $0.75 Out: $4.50
65	Grok Build 0.1 0616 SpaceXAI	39.8	94 t/s	1ms	256k	39.8	39.8	39.8	In: $1.00 Out: $2.00
66	Qwen3.6 Plus Alibaba	39.6	56 t/s	2ms	1M	39.6	39.6	39.6	In: $0.50 Out: $3.00
67	Gemini 3 Pro Preview (high) Google	39.6	—	—	1M	60.0	59.4	58.8	In: $2.00 Out: $12.00
68	Z GLM-5 (Reasoning)Open Z AI	39.5	40 t/s	2ms	200k	39.5	39.5	39.5	In: $1.00 Out: $3.20
69	GPT-5.4 (low) OpenAI	39.1	111 t/s	2ms	1.1M	39.1	39.1	39.1	In: $2.50 Out: $15.00
70	Qwen3.7 Plus Alibaba	39.0	53 t/s	2ms	1M	39.0	39.0	39.0	In: $0.40 Out: $1.60
71	C JT-4.1 Flash 236B A21B China Mobile	38.8	—	—	256k	38.8	38.8	38.8	In: — Out: —
72	S Agnes 2.5 Pro Alpha Sapiens AI	38.8	131 t/s	153ms	1M	38.8	38.8	38.8	In: $0.45 Out: $0.90
73	GPT-5.4 nano (xhigh) OpenAI	38.2	167 t/s	5ms	400k	38.2	38.2	38.2	In: $0.20 Out: $1.25
74	MiniMax-M2.7Open MiniMax	38.1	106 t/s	2ms	205k	38.1	38.1	38.1	In: $0.30 Out: $1.20
75	Z GLM-5-Turbo Z AI	38.1	—	—	200k	38.1	38.1	38.1	In: — Out: —
76	GPT-5.6 Luna (medium) OpenAI	38.1	156 t/s	2ms	1M	38.1	38.1	38.1	In: $0.20 Out: $1.20
77	GPT-5.2 (medium) OpenAI	38.0	—	—	400k	38.0	38.0	38.0	In: $1.75 Out: $14.00
78	Claude Opus 4.6 (Non-reasoning, High Effort) Anthropic	37.8	37 t/s	3ms	1M	37.8	37.8	37.8	In: $5.00 Out: $25.00
79	Nemotron 3 Ultra 550B A55B (Reasoning)Open NVIDIA	37.8	135 t/s	2ms	262k	37.8	37.8	37.8	In: $0.60 Out: $2.75
80	Gemini 3 Flash Preview (Reasoning) Google	37.8	179 t/s	7ms	1M	37.8	37.8	37.8	In: $0.50 Out: $3.00
81	Grok 4.3 (high) SpaceXAI	37.6	141 t/s	20ms	1M	37.6	37.6	37.6	In: $1.25 Out: $2.50
82	DeepSeek V4 Flash (Reasoning, High Effort)Open DeepSeek	37.5	—	—	1M	37.5	37.5	37.5	In: $0.14 Out: $0.28
83	MiMo-V2.5Open Xiaomi	37.2	81 t/s	3ms	1M	37.2	37.2	37.2	In: $0.14 Out: $0.28
84	Qwen3.6 27B (Reasoning)Open Alibaba	37.1	59 t/s	4ms	262k	37.1	37.1	37.1	In: $0.60 Out: $3.60
85	Grok 4.20 0309 v2 (Reasoning) SpaceXAI	37.0	181 t/s	14ms	2M	37.0	37.0	37.0	In: $1.25 Out: $2.50
86	GPT-5.1 (high) OpenAI	36.9	106 t/s	47ms	272k	36.9	36.9	36.9	In: $1.25 Out: $10.00
87	Grok 4.20 0309 (Reasoning) SpaceXAI	36.5	—	—	2M	36.5	36.5	36.5	In: $2.00 Out: $6.00
88	Gemini 3.5 Flash-Lite Google	36.5	391 t/s	10ms	1M	36.5	36.5	36.5	In: $0.30 Out: $2.50
89	Claude 4.5 Sonnet (Reasoning) Anthropic	36.4	47 t/s	12ms	1M	36.4	36.4	36.4	In: $3.00 Out: $15.00
90	MiMo-V2-Omni-0327 Xiaomi	36.4	—	—	256k	36.4	36.4	36.4	In: — Out: —
91	GPT-5 Codex (high) OpenAI	36.1	—	—	400k	36.1	36.1	36.1	In: $1.25 Out: $10.00
92	Grok 4.3 (medium) SpaceXAI	36.0	140 t/s	11ms	1M	36.0	36.0	36.0	In: $1.25 Out: $2.50
93	Claude Sonnet 4.6 (Non-reasoning, High Effort) Anthropic	35.9	43 t/s	1ms	1M	35.9	35.9	35.9	In: $3.00 Out: $15.00
94	Grok 4.3 (low) SpaceXAI	35.4	128 t/s	7ms	1M	35.4	35.4	35.4	In: $1.25 Out: $2.50
95	GPT-5.5 (Non-reasoning) OpenAI	35.4	68 t/s	1ms	922k	35.4	35.4	35.4	In: $5.00 Out: $30.00
96	Kimi K2.5 (Reasoning)Open Kimi	35.4	52 t/s	3ms	256k	35.4	35.4	35.4	In: $0.60 Out: $3.00
97	Z GLM-5.1 (Non-reasoning)Open Z AI	35.4	64 t/s	2ms	200k	35.4	35.4	35.4	In: $1.38 Out: $4.40
98	MiMo-V2-Omni Xiaomi	35.0	—	—	256k	35.0	35.0	35.0	In: — Out: —
99	Gemini 3.5 Flash (minimal) Google	34.9	170 t/s	1ms	1M	34.9	34.9	34.9	In: $1.50 Out: $9.00
100	GPT-5.1 Codex (high) OpenAI	34.7	—	—	400k	34.7	34.7	34.7	In: $1.25 Out: $10.00
101	GPT-5 (high) OpenAI	34.7	109 t/s	86ms	400k	34.7	34.7	34.7	In: $1.25 Out: $10.00
102	Claude Opus 4.5 (Non-reasoning) Anthropic	34.7	46 t/s	1ms	200k	34.7	34.7	34.7	In: $5.00 Out: $25.00
103	Kimi K2.6 (Non-reasoning)Open Kimi	34.6	32 t/s	3ms	256k	34.6	34.6	34.6	In: $0.95 Out: $4.00
104	Z GLM 5V Turbo (Reasoning) Z AI	34.5	—	—	200k	34.5	34.5	34.5	In: — Out: —
105	Claude Sonnet 4.6 (Non-reasoning, Low Effort) Anthropic	34.3	44 t/s	1ms	1M	34.3	34.3	34.3	In: $3.00 Out: $15.00
106	Z GLM-5.2 (Non-reasoning)Open Z AI	34.1	108 t/s	2ms	1M	34.1	34.1	34.1	In: $1.35 Out: $4.25
107	GPT-5.6 Terra (Non-reasoning) OpenAI	34.0	115 t/s	1ms	1M	34.0	34.0	34.0	In: $2.00 Out: $12.00
108	Qwen3.5 27B (Reasoning)Open Alibaba	33.8	77 t/s	6ms	262k	33.8	33.8	33.8	In: $0.30 Out: $2.40
109	GPT-5 (medium) OpenAI	33.7	90 t/s	32ms	400k	33.7	33.7	33.7	In: $1.25 Out: $10.00
110	Claude 4.1 Opus (Reasoning) Anthropic	33.7	—	—	200k	33.7	33.7	33.7	In: $15.00 Out: $75.00
111	Z GLM-4.7 (Reasoning)Open Z AI	33.7	104 t/s	1ms	200k	33.7	33.7	33.7	In: $0.60 Out: $2.20
112	K KAT Coder Pro V2 KwaiKAT	33.7	107 t/s	2ms	256k	33.7	33.7	33.7	In: $0.30 Out: $1.20
113	Qwen3.5 397B A17B (Reasoning)Open Alibaba	33.7	68 t/s	2ms	262k	33.7	33.7	33.7	In: $0.60 Out: $3.60
114	MiniMax-M2.5Open MiniMax	33.7	98 t/s	2ms	205k	33.7	33.7	33.7	In: $0.30 Out: $1.20
115	T Hy3-preview (Reasoning)Open Tencent	33.6	—	—	256k	33.6	33.6	33.6	In: $0.06 Out: $0.21
116	GPT-5.5 Instant (May 2026) OpenAI	33.5	—	—	400k	33.5	33.5	33.5	In: $5.00 Out: $30.00
117	L LongCat 2.0Open LongCat	33.5	42 t/s	3ms	1M	33.5	33.5	33.5	In: $0.75 Out: $2.95
118	Grok 4 SpaceXAI	33.3	—	—	256k	33.3	33.3	33.3	In: $3.00 Out: $15.00
119	GPT-5.6 Luna (low) OpenAI	33.3	152 t/s	2ms	1M	33.3	33.3	33.3	In: $0.20 Out: $1.20
120	MiMo-V2-Flash (Feb 2026)Open Xiaomi	33.2	—	—	256k	33.2	33.2	33.2	In: — Out: —
121	Gemini 3 Pro Preview (low) Google	33.1	—	—	1M	33.1	33.1	33.1	In: $2.00 Out: $12.00
122	Kimi K2 ThinkingOpen Kimi	32.7	123 t/s	1ms	256k	32.7	32.7	32.7	In: $0.60 Out: $2.50
123	o3-pro OpenAI	32.5	—	—	200k	32.5	32.5	32.5	In: $20.00 Out: $80.00
124	Z GLM-5 (Non-reasoning)Open Z AI	32.4	40 t/s	2ms	200k	32.4	32.4	32.4	In: $1.00 Out: $3.20
125	Qwen3.5 122B A10B (Reasoning)Open Alibaba	32.3	133 t/s	2ms	262k	32.3	32.3	32.3	In: $0.40 Out: $3.20
126	DeepSeek V3.2 (Reasoning)Open DeepSeek	32.0	—	—	128k	32.0	32.0	32.0	In: $0.28 Out: $0.42
127	Qwen3.5 397B A17B (Non-reasoning)Open Alibaba	32.0	65 t/s	2ms	262k	32.0	32.0	32.0	In: $0.60 Out: $3.60
128	Qwen3 Max Thinking Alibaba	31.7	—	—	256k	31.7	31.7	31.7	In: — Out: —
129	Qwen3.6 35B A3B (Reasoning)Open Alibaba	31.6	152 t/s	2ms	262k	31.6	31.6	31.6	In: $0.25 Out: $1.49
130	MiniMax-M2.1Open MiniMax	31.4	136 t/s	2ms	205k	31.4	31.4	31.4	In: $0.30 Out: $1.20
131	DeepSeek V4 Pro (Non-reasoning)Open DeepSeek	31.2	64 t/s	2ms	1M	31.2	31.2	31.2	In: $0.43 Out: $0.87
132	MiMo-V2-Flash (Reasoning)Open Xiaomi	31.2	—	—	256k	31.2	31.2	31.2	In: $0.10 Out: $0.30
133	GPT-5 (low) OpenAI	31.2	84 t/s	8ms	400k	31.2	31.2	31.2	In: $1.25 Out: $10.00
134	Claude 4 Opus (Reasoning) Anthropic	31.0	—	—	200k	31.0	31.0	31.0	In: $15.00 Out: $75.00
135	A G9v3-39A5BOpen AI9Stars	30.9	—	—	131k	30.9	30.9	30.9	In: $0.00 Out: $0.00
136	GPT-5 mini (medium) OpenAI	30.9	97 t/s	14ms	400k	30.9	30.9	30.9	In: $0.25 Out: $2.00
137	Qwen3.5 Omni Plus Alibaba	30.6	53 t/s	2ms	256k	30.6	30.6	30.6	In: $0.40 Out: $4.80
138	GPT-5.1 Codex mini (high) OpenAI	30.6	—	—	400k	30.6	30.6	30.6	In: $0.25 Out: $2.00
139	Grok 4.1 Fast (Reasoning) SpaceXAI	30.6	—	—	2M	30.6	30.6	30.6	In: — Out: —
140	I Ring-2.6-1TOpen InclusionAI	30.6	—	—	262k	30.6	30.6	30.6	In: $0.30 Out: $2.50
141	Qwen3.6 27B (Non-reasoning)Open Alibaba	30.5	59 t/s	4ms	262k	30.5	30.5	30.5	In: $0.60 Out: $3.60
142	DeepSeek V3.1 Terminus (Reasoning)Open DeepSeek	30.4	—	—	128k	30.4	30.4	30.4	In: $1.64 Out: $2.75
143	o3 OpenAI	30.4	127 t/s	7ms	200k	30.4	30.4	30.4	In: $2.00 Out: $8.00
144	S Step 3.7 FlashOpen StepFun	30.3	395 t/s	1ms	262k	30.3	30.3	30.3	In: $0.20 Out: $1.15
145	GPT-5.4 nano (medium) OpenAI	30.2	165 t/s	5ms	400k	30.2	30.2	30.2	In: $0.20 Out: $1.25
146	Mistral Medium 3.5Open Mistral	29.9	143 t/s	2ms	256k	29.9	29.9	29.9	In: $1.50 Out: $7.50
147	GPT-5.4 mini (medium) OpenAI	29.8	189 t/s	13ms	400k	29.8	29.8	29.8	In: $0.75 Out: $4.50
148	Claude 4.5 Haiku (Reasoning) Anthropic	29.6	98 t/s	20ms	200k	29.6	29.6	29.6	In: $1.00 Out: $5.00
149	Kimi K2.5 (Non-reasoning)Open Kimi	29.4	52 t/s	3ms	256k	29.4	29.4	29.4	In: $0.60 Out: $3.00
150	Gemma 4 31B (Reasoning)Open Google	29.4	35 t/s	1ms	256k	29.4	29.4	29.4	In: $0.00 Out: $0.00
151	Qwen3.5 27B (Non-reasoning)Open Alibaba	29.3	83 t/s	6ms	262k	29.3	29.3	29.3	In: $0.30 Out: $2.40
152	Claude 4.5 Sonnet (Non-reasoning) Anthropic	29.3	41 t/s	2ms	1M	29.3	29.3	29.3	In: $3.00 Out: $15.00
153	Qwen3.5 35B A3B (Reasoning)Open Alibaba	29.3	156 t/s	2ms	262k	29.3	29.3	29.3	In: $0.25 Out: $2.00
154	Claude 4 Sonnet (Reasoning) Anthropic	28.9	—	—	1M	28.9	28.9	28.9	In: $3.00 Out: $15.00
155	GPT-5.5 Instant (June 2026) OpenAI	28.9	170 t/s	1ms	400k	28.9	28.9	28.9	In: $5.00 Out: $30.00
156	Z GLM-4.6 (Reasoning)Open Z AI	28.7	49 t/s	2ms	200k	28.7	28.7	28.7	In: $0.55 Out: $2.20
157	DeepSeek V4 Flash (Non-reasoning)Open DeepSeek	28.7	121 t/s	1ms	1M	28.7	28.7	28.7	In: $0.14 Out: $0.28
158	C JT-35B-Flash China Mobile	28.4	—	—	256k	28.4	28.4	28.4	In: — Out: —
159	MiniMax-M2Open MiniMax	28.3	101 t/s	2ms	205k	28.3	28.3	28.3	In: $0.30 Out: $1.20
160	K KAT-Coder-Pro V1 KwaiKAT	28.3	—	—	256k	28.3	28.3	28.3	In: — Out: —
161	Claude 4.1 Opus (Non-reasoning) Anthropic	28.2	—	—	200k	28.2	28.2	28.2	In: $15.00 Out: $75.00
162	MiMo-V2.5-Pro (Non-reasoning)Open Xiaomi	27.9	45 t/s	3ms	1M	27.9	27.9	27.9	In: $0.43 Out: $0.87
163	GPT-5.4 (Non-reasoning) OpenAI	27.7	106 t/s	1ms	1.1M	27.7	27.7	27.7	In: $2.50 Out: $15.00
164	Qwen3.5 122B A10B (Non-reasoning)Open Alibaba	27.6	148 t/s	2ms	262k	27.6	27.6	27.6	In: $0.40 Out: $3.20
165	Grok 4 Fast (Reasoning) SpaceXAI	27.4	—	—	2M	27.4	27.4	27.4	In: $0.20 Out: $0.50
166	Gemini 3 Flash Preview (Non-reasoning) Google	27.4	185 t/s	1ms	1M	27.4	27.4	27.4	In: $0.50 Out: $3.00
167	Claude 3.7 Sonnet (Reasoning) Anthropic	27.1	—	—	200k	27.1	27.1	27.1	In: — Out: —
168	GPT-5.6 Luna (Non-reasoning) OpenAI	26.6	164 t/s	1ms	1M	26.6	26.6	26.6	In: $0.20 Out: $1.20
169	Z GLM-4.7 (Non-reasoning)Open Z AI	26.6	101 t/s	1ms	200k	26.6	26.6	26.6	In: $0.60 Out: $2.20
170	T Hy3-preview (Non-reasoning)Open Tencent	26.1	—	—	256k	26.1	26.1	26.1	In: $0.06 Out: $0.21
171	I Ling-2.6-1TOpen InclusionAI	26.1	—	—	262k	26.1	26.1	26.1	In: $0.30 Out: $2.50
172	GPT-5.2 (Non-reasoning) OpenAI	26.0	73 t/s	1ms	400k	26.0	26.0	26.0	In: $1.75 Out: $14.00
173	S Step 3.5 Flash 2603 StepFun	26.0	155 t/s	1ms	256k	26.0	26.0	26.0	In: $0.10 Out: $0.30
174	Doubao Seed Code ByteDance Seed	26.0	—	—	256k	26.0	26.0	26.0	In: — Out: —
175	Gemini 2.5 Pro Google	25.8	143 t/s	23ms	1M	25.8	25.8	25.8	In: $1.25 Out: $10.00
176	Gemma 4 26B A4B (Reasoning)Open Google	25.7	—	—	256k	25.7	25.7	25.7	In: $0.13 Out: $0.40
177	o4-mini (high) OpenAI	25.6	168 t/s	20ms	200k	25.6	25.6	25.6	In: $1.10 Out: $4.40
178	S Step 3.5 FlashOpen StepFun	25.5	146 t/s	1ms	256k	25.5	25.5	25.5	In: $0.10 Out: $0.30
179	Claude 4 Opus (Non-reasoning) Anthropic	25.5	—	—	200k	25.5	25.5	25.5	In: $15.00 Out: $75.00
180	Claude 4 Sonnet (Non-reasoning) Anthropic	25.5	—	—	1M	25.5	25.5	25.5	In: $3.00 Out: $15.00
181	DeepSeek V3.2 Exp (Reasoning)Open DeepSeek	25.4	—	—	128k	25.4	25.4	25.4	In: $0.28 Out: $0.42
182	NVIDIA Nemotron 3 Super 120B A12B (Reasoning)Open NVIDIA	25.4	143 t/s	2ms	1M	25.4	25.4	25.4	In: $0.20 Out: $0.80
183	GPT-5 mini (high) OpenAI	25.3	93 t/s	97ms	400k	25.3	25.3	25.3	In: $0.25 Out: $2.00
184	Gemini 3.1 Flash-Lite Google	25.0	312 t/s	6ms	1M	25.0	25.0	25.0	In: $0.25 Out: $1.50
185	Qwen3 Max Thinking (Preview) Alibaba	25.0	60 t/s	4ms	262k	25.0	25.0	25.0	In: $1.20 Out: $6.00
186	Grok 4.3 (Non-reasoning) SpaceXAI	24.8	134 t/s	1ms	1M	24.8	24.8	24.8	In: $1.25 Out: $2.50
187	MiMo-V2-Flash (Non-reasoning)Open Xiaomi	24.7	—	—	256k	24.7	24.7	24.7	In: — Out: —
188	DeepSeek V3.2 (Non-reasoning)Open DeepSeek	24.7	—	—	128k	24.7	24.7	24.7	In: $0.28 Out: $0.42
189	Qwen3.6 35B A3B (Non-reasoning)Open Alibaba	24.2	166 t/s	2ms	262k	24.2	24.2	24.2	In: $0.38 Out: $2.25
190	Qwen3 Max Alibaba	24.0	64 t/s	2ms	262k	24.0	24.0	24.0	In: $1.20 Out: $6.00
191	Qwen3.5 35B A3B (Non-reasoning)Open Alibaba	24.0	175 t/s	2ms	262k	24.0	24.0	24.0	In: $0.25 Out: $2.00
192	gpt-oss-120b (high)Open OpenAI	23.8	199 t/s	1ms	131k	23.8	23.8	23.8	In: $0.15 Out: $0.60
193	Gemini 2.5 Flash Preview (Sep '25) (Reasoning) Google	23.8	—	—	1M	23.8	23.8	23.8	In: — Out: —
194	Claude 4.5 Haiku (Non-reasoning) Anthropic	23.7	90 t/s	1ms	200k	23.7	23.7	23.7	In: $1.00 Out: $5.00
195	Kimi K2 0905Open Kimi	23.5	37 t/s	1ms	256k	23.5	23.5	23.5	In: $0.60 Out: $2.50
196	Claude 3.7 Sonnet (Non-reasoning) Anthropic	23.5	—	—	200k	23.5	23.5	23.5	In: $3.00 Out: $15.00
197	o1 OpenAI	23.4	—	—	200k	23.4	23.4	23.4	In: $15.00 Out: $60.00
198	Gemini 2.5 Pro Preview (Mar' 25) Google	23.0	—	—	1M	23.0	23.0	23.0	In: — Out: —
199	Z GLM-4.6 (Non-reasoning)Open Z AI	23.0	59 t/s	3ms	200k	23.0	23.0	23.0	In: $0.57 Out: $2.20
200	Z GLM-4.7-Flash (Reasoning)Open Z AI	22.9	98 t/s	1ms	200k	22.9	22.9	22.9	In: $0.07 Out: $0.40
201	C Command A+Open Cohere	22.5	198 t/s	0ms	192k	22.5	22.5	22.5	In: $0.00 Out: $0.00
202	Grok 3 mini Reasoning (high) SpaceXAI	22.5	81 t/s	1ms	1M	22.5	22.5	22.5	In: $0.30 Out: $0.50
203	Grok 4.20 0309 (Non-reasoning) SpaceXAI	22.5	—	—	2M	22.5	22.5	22.5	In: $2.00 Out: $6.00
204	Gemini 2.5 Pro Preview (May' 25) Google	22.3	—	—	1M	22.3	22.3	22.3	In: $1.25 Out: $10.00
205	DeepSeek V3.2 SpecialeOpen DeepSeek	22.2	—	—	128k	22.2	22.2	22.2	In: — Out: —
206	L K-EXAONE (Reasoning)Open LG AI Research	22.1	—	—	256k	22.1	22.1	22.1	In: — Out: —
207	ERNIE 5.0 Thinking Preview Baidu	21.9	—	—	128k	21.9	21.9	21.9	In: — Out: —
208	Grok 4.20 0309 v2 (Non-reasoning) SpaceXAI	21.8	168 t/s	1ms	2M	21.8	21.8	21.8	In: $1.25 Out: $2.50
209	Gemma 4 12B (Reasoning)Open Google	21.8	111 t/s	2ms	256k	21.8	21.8	21.8	In: $0.10 Out: $0.30
210	Gemma 4 31B (Non-reasoning)Open Google	21.8	68 t/s	2ms	256k	21.8	21.8	21.8	In: $0.14 Out: $0.40
211	A Nova 2.0 Pro Preview (medium) Amazon	21.8	120 t/s	13ms	256k	21.8	21.8	21.8	In: $1.25 Out: $10.00
212	Grok Code Fast 1 SpaceXAI	21.6	—	—	256k	21.6	21.6	21.6	In: — Out: —
213	Qwen3.5 9B (Reasoning)Open Alibaba	21.4	83 t/s	1ms	262k	21.4	21.4	21.4	In: $0.14 Out: $0.20
214	I Mercury 2 Inception	21.4	788 t/s	4ms	128k	21.4	21.4	21.4	In: $0.25 Out: $0.75
215	DeepSeek V3.1 Terminus (Non-reasoning)Open DeepSeek	21.4	—	—	128k	21.4	21.4	21.4	In: $0.27 Out: $1.00
216	DeepSeek V3.2 Exp (Non-reasoning)Open DeepSeek	21.3	—	—	128k	21.3	21.3	21.3	In: $0.28 Out: $0.42
217	S Apriel-v1.5-15B-ThinkerOpen ServiceNow	21.2	—	—	128k	21.2	21.2	21.2	In: $0.00 Out: $0.00
218	Qwen3 Coder NextOpen Alibaba	21.1	133 t/s	2ms	256k	21.1	21.1	21.1	In: $0.35 Out: $1.20
219	DeepSeek V3.1 (Non-reasoning)Open DeepSeek	21.0	—	—	128k	21.0	21.0	21.0	In: $0.56 Out: $1.68
220	A Nova 2.0 Omni (medium) Amazon	20.9	—	—	1M	20.9	20.9	20.9	In: $0.30 Out: $2.50
221	DeepSeek V3.1 (Reasoning)Open DeepSeek	20.7	—	—	128k	20.7	20.7	20.7	In: $0.59 Out: $1.69
222	Qwen3 VL 235B A22B (Reasoning)Open Alibaba	20.6	58 t/s	3ms	262k	20.6	20.6	20.6	In: $0.70 Out: $8.40
223	S Apriel-v1.6-15B-ThinkerOpen ServiceNow	20.5	—	—	128k	20.5	20.5	20.5	In: $0.00 Out: $0.00
224	GPT-5.1 (Non-reasoning) OpenAI	20.4	102 t/s	1ms	400k	20.4	20.4	20.4	In: $1.25 Out: $10.00
225	Qwen3.5 9B (Non-reasoning)Open Alibaba	20.3	—	—	262k	20.3	20.3	20.3	In: — Out: —
226	L EXAONE 4.5 33BOpen LG AI Research	20.2	—	—	262k	20.2	20.2	20.2	In: — Out: —
227	Gemma 4 26B A4B (Non-reasoning)Open Google	20.1	65 t/s	1ms	256k	20.1	20.1	20.1	In: $0.13 Out: $0.40
228	Qwen3.5 4B (Reasoning)Open Alibaba	20.1	45 t/s	1ms	262k	20.1	20.1	20.1	In: $0.03 Out: $0.15
229	DeepSeek R1 0528 (May '25)Open DeepSeek	20.1	—	—	128k	20.1	20.1	20.1	In: $1.35 Out: $4.20
230	Gemini 2.5 Flash (Reasoning) Google	20.1	231 t/s	15ms	1M	20.1	20.1	20.1	In: $0.30 Out: $2.50
231	GPT-5 nano (high) OpenAI	19.9	150 t/s	120ms	400k	19.9	19.9	19.9	In: $0.05 Out: $0.40
232	C North Mini CodeOpen Cohere	19.8	22 t/s	1ms	256k	19.8	19.8	19.8	In: $0.00 Out: $0.00
233	Qwen3 235B A22B 2507 (Reasoning)Open Alibaba	19.6	55 t/s	3ms	256k	19.6	19.6	19.6	In: $0.70 Out: $8.40
234	A Nova 2.0 Pro Preview (low) Amazon	19.6	126 t/s	10ms	256k	19.6	19.6	19.6	In: $1.25 Out: $10.00
235	Mistral Small 4 (Reasoning)Open Mistral	19.6	160 t/s	1ms	256k	19.6	19.6	19.6	In: $0.15 Out: $0.60
236	Z GLM-4.5 (Reasoning)Open Z AI	19.5	—	—	128k	19.5	19.5	19.5	In: — Out: —
237	Kimi K2Open Kimi	19.4	38 t/s	1ms	128k	19.4	19.4	19.4	In: $0.57 Out: $2.30
238	GPT-4.1 OpenAI	19.4	151 t/s	1ms	1M	19.4	19.4	19.4	In: $2.00 Out: $8.00
239	Qwen3 Max (Preview) Alibaba	19.2	59 t/s	4ms	262k	19.2	19.2	19.2	In: $1.20 Out: $6.00
240	Devstral 2Open Mistral	19.2	44 t/s	1ms	256k	19.2	19.2	19.2	In: $0.00 Out: $0.00
241	A Nova 2.0 Lite (medium) Amazon	19.0	162 t/s	17ms	1M	19.0	19.0	19.0	In: $0.30 Out: $2.50
242	GPT-5 nano (medium) OpenAI	19.0	148 t/s	48ms	400k	19.0	19.0	19.0	In: $0.05 Out: $0.40
243	Qwen3.5 Omni Flash Alibaba	19.0	249 t/s	2ms	256k	19.0	19.0	19.0	In: $0.10 Out: $0.80
244	o3-mini OpenAI	19.0	215 t/s	5ms	200k	19.0	19.0	19.0	In: $1.10 Out: $4.40
245	o1-pro OpenAI	18.9	—	—	200k	18.9	18.9	18.9	In: $150.00 Out: $600.00
246	Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning) Google	18.8	—	—	1M	18.8	18.8	18.8	In: — Out: —
247	C JT-MINI China Mobile	18.5	—	—	128k	18.5	18.5	18.5	In: — Out: —
248	DeepSeek R1 (Jan '25)Open DeepSeek	18.5	—	—	128k	18.5	18.5	18.5	In: $1.68 Out: $4.70
249	Grok 3 SpaceXAI	18.4	—	—	1M	18.4	18.4	18.4	In: $4.00 Out: $20.00
250	Seed-OSS-36B-InstructOpen ByteDance Seed	18.3	35 t/s	3ms	512k	18.3	18.3	18.3	In: $0.21 Out: $0.57
251	A Nova 2.0 Lite (high) Amazon	18.2	141 t/s	19ms	1M	18.2	18.2	18.2	In: $0.30 Out: $2.50
252	A Trinity Large ThinkingOpen Arcee AI	18.2	191 t/s	1ms	512k	18.2	18.2	18.2	In: $0.23 Out: $0.88
253	Qwen3 235B A22B 2507 InstructOpen Alibaba	18.2	57 t/s	2ms	256k	18.2	18.2	18.2	In: $0.70 Out: $2.80
254	Qwen3 Coder 480B A35B InstructOpen Alibaba	18.0	66 t/s	3ms	262k	18.0	18.0	18.0	In: $1.50 Out: $7.50
255	Qwen3 VL 32B (Reasoning)Open Alibaba	17.9	90 t/s	3ms	256k	17.9	17.9	17.9	In: $0.70 Out: $8.40
256	Magistral Medium 1.2 Mistral	17.9	114 t/s	2ms	128k	17.9	17.9	17.9	In: $2.00 Out: $5.00
257	P Sonar Reasoning Pro Perplexity	17.8	—	—	127k	17.8	17.8	17.8	In: — Out: —
258	A Nova 2.0 Lite (low) Amazon	17.8	157 t/s	13ms	1M	17.8	17.8	17.8	In: $0.30 Out: $2.50
259	M HyperNova 60B 2605Open Multiverse Computing	17.8	347 t/s	1ms	131k	17.8	17.8	17.8	In: $0.04 Out: $0.14
260	MiniMax M1 80kOpen MiniMax	17.7	—	—	1M	17.7	17.7	17.7	In: $0.55 Out: $2.20
261	GPT-5.4 nano (Non-Reasoning) OpenAI	17.6	165 t/s	1ms	400k	17.6	17.6	17.6	In: $0.20 Out: $1.25
262	Nemotron Cascade 2 30B A3BOpen NVIDIA	17.6	—	—	1M	17.6	17.6	17.6	In: — Out: —
263	Gemini 2.5 Flash Preview (Reasoning) Google	17.5	—	—	1M	17.5	17.5	17.5	In: — Out: —
264	Devstral Small 2Open Mistral	17.4	74 t/s	2ms	256k	17.4	17.4	17.4	In: $0.00 Out: $0.00
265	M K2 Think V2Open MBZUAI Institute of Foundation Models	17.3	—	—	262k	17.3	17.3	17.3	In: — Out: —
266	L LongCat Flash LiteOpen LongCat	17.2	—	—	256k	17.2	17.2	17.2	In: — Out: —
267	GPT-5 (minimal) OpenAI	17.2	78 t/s	1ms	400k	17.2	17.2	17.2	In: $1.25 Out: $10.00
268	o1-preview OpenAI	17.0	—	—	128k	17.0	17.0	17.0	In: $16.50 Out: $66.00
269	N HyperCLOVA X SEED Think (32B)Open Naver	17.0	—	—	128k	17.0	17.0	17.0	In: — Out: —
270	Grok 4.1 Fast (Non-reasoning) SpaceXAI	16.9	—	—	2M	16.9	16.9	16.9	In: — Out: —
271	Z GLM-4.6V (Reasoning)Open Z AI	16.8	77 t/s	4ms	128k	16.8	16.8	16.8	In: $0.30 Out: $0.90
272	L K-EXAONE (Non-reasoning)Open LG AI Research	16.7	—	—	256k	16.7	16.7	16.7	In: — Out: —
273	Qwen3 Next 80B A3B (Reasoning)Open Alibaba	16.7	208 t/s	2ms	262k	16.7	16.7	16.7	In: $0.50 Out: $6.00
274	GPT-5.4 mini (Non-Reasoning) OpenAI	16.6	179 t/s	1ms	400k	16.6	16.6	16.6	In: $0.75 Out: $4.50
275	A Nova 2.0 Omni (low) Amazon	16.6	—	—	1M	16.6	16.6	16.6	In: $0.30 Out: $2.50
276	Z GLM-4.5-AirOpen Z AI	16.5	86 t/s	3ms	128k	16.5	16.5	16.5	In: $0.17 Out: $0.98
277	Grok 4 Fast (Non-reasoning) SpaceXAI	16.5	—	—	2M	16.5	16.5	16.5	In: $0.20 Out: $0.50
278	K Mi:dm K 2.5 Pro Korea Telecom	16.4	—	—	128k	16.4	16.4	16.4	In: — Out: —
279	I Ring-1TOpen InclusionAI	16.2	—	—	128k	16.2	16.2	16.2	In: — Out: —
280	A G9v3-3BOpen AI9Stars	16.1	—	—	131k	16.1	16.1	16.1	In: $0.00 Out: $0.00
281	Qwen3.5 4B (Non-reasoning)Open Alibaba	16.0	25 t/s	1ms	262k	16.0	16.0	16.0	In: $0.03 Out: $0.15
282	Mistral Large 3Open Mistral	15.9	47 t/s	1ms	256k	15.9	15.9	15.9	In: $0.50 Out: $1.50
283	o3-mini (high) OpenAI	15.6	229 t/s	22ms	200k	15.6	15.6	15.6	In: $1.10 Out: $4.40
284	P INTELLECT-3Open Prime Intellect	15.6	—	—	131k	15.6	15.6	15.6	In: — Out: —
285	Z GLM-4.7-Flash (Non-reasoning)Open Z AI	15.5	118 t/s	2ms	200k	15.5	15.5	15.5	In: $0.07 Out: $0.40
286	DeepSeek V3 0324Open DeepSeek	15.4	—	—	128k	15.4	15.4	15.4	In: $0.27 Out: $1.12
287	GPT-5 (ChatGPT) OpenAI	15.3	—	—	128k	15.3	15.3	15.3	In: — Out: —
288	U Solar Open 100B (Reasoning)Open Upstage	15.1	—	—	128k	15.1	15.1	15.1	In: — Out: —
289	Grok 3 Reasoning Beta SpaceXAI	15.1	—	—	1M	15.1	15.1	15.1	In: — Out: —
290	Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning) Google	15.1	—	—	1M	15.1	15.1	15.1	In: $0.10 Out: $0.40
291	Nemotron 3 Nano Omni 30B A3B ReasoningOpen NVIDIA	14.9	323 t/s	1ms	256k	14.9	14.9	14.9	In: $0.07 Out: $0.30
292	gpt-oss-120b (low)Open OpenAI	14.9	266 t/s	1ms	131k	14.9	14.9	14.9	In: $0.15 Out: $0.59
293	gpt-oss-20b (high)Open OpenAI	14.9	210 t/s	1ms	131k	14.9	14.9	14.9	In: $0.07 Out: $0.20
294	GPT-4.1 mini OpenAI	14.8	76 t/s	1ms	1M	14.8	14.8	14.8	In: $0.40 Out: $1.60
295	Mistral Small 3.1Open Mistral	14.7	149 t/s	1ms	128k	14.7	14.7	14.7	In: $0.10 Out: $0.30
296	Mistral Medium 3.1 Mistral	14.7	84 t/s	2ms	128k	14.7	14.7	14.7	In: $0.40 Out: $2.00
297	Qwen3 30B A3B 2507 (Reasoning)Open Alibaba	14.4	137 t/s	3ms	262k	14.4	14.4	14.4	In: $0.20 Out: $2.40
298	MiniMax M1 40kOpen MiniMax	14.4	—	—	1M	14.4	14.4	14.4	In: — Out: —
299	A Nova 2.0 Pro Preview (Non-reasoning) Amazon	14.4	103 t/s	1ms	256k	14.4	14.4	14.4	In: $1.25 Out: $10.00
300	gpt-oss-20b (low)Open OpenAI	14.3	216 t/s	1ms	131k	14.3	14.3	14.3	In: $0.07 Out: $0.23
301	Qwen3 VL 235B A22B InstructOpen Alibaba	14.3	53 t/s	3ms	262k	14.3	14.3	14.3	In: $0.70 Out: $2.80
302	Llama 4 MaverickOpen Meta	14.3	113 t/s	1ms	1M	14.3	14.3	14.3	In: $0.27 Out: $0.91
303	GPT-5 mini (minimal) OpenAI	14.3	100 t/s	1ms	400k	14.3	14.3	14.3	In: $0.25 Out: $2.00
304	M K2-V2 (high)Open MBZUAI Institute of Foundation Models	14.2	—	—	512k	14.2	14.2	14.2	In: — Out: —
305	NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)Open NVIDIA	14.2	263 t/s	1ms	1M	14.2	14.2	14.2	In: $0.05 Out: $0.20
306	DeepSeek V3 (Dec '24)Open DeepSeek	14.2	—	—	128k	14.2	14.2	14.2	In: $0.36 Out: $0.89
307	Gemini 2.5 Flash (Non-reasoning) Google	14.1	218 t/s	0ms	1M	14.1	14.1	14.1	In: $0.30 Out: $2.50
308	U Solar Pro 3 Upstage	14.1	121 t/s	2ms	128k	14.1	14.1	14.1	In: $0.15 Out: $0.60
309	I Ling 2.6 FlashOpen InclusionAI	14.1	91 t/s	1ms	262k	14.1	14.1	14.1	In: $0.10 Out: $0.30
310	o1-mini OpenAI	14.0	—	—	128k	14.0	14.0	14.0	In: — Out: —
311	Qwen3 Next 80B A3B InstructOpen Alibaba	13.7	185 t/s	2ms	262k	13.7	13.7	13.7	In: $0.50 Out: $2.00
312	T Tri-21B-think PreviewOpen Trillion Labs	13.6	—	—	32k	13.6	13.6	13.6	In: — Out: —
313	Qwen3 Coder 30B A3B InstructOpen Alibaba	13.6	94 t/s	3ms	262k	13.6	13.6	13.6	In: $0.45 Out: $2.25
314	GPT-4.5 (Preview) OpenAI	13.6	—	—	128k	13.6	13.6	13.6	In: — Out: —
315	DiffusionGemma 26B A4BOpen Google	13.5	—	—	256k	13.5	13.5	13.5	In: — Out: —
316	Qwen3 235B A22B (Reasoning)Open Alibaba	13.4	61 t/s	3ms	33k	13.4	13.4	13.4	In: $0.70 Out: $8.40
317	QwQ 32BOpen Alibaba	13.4	—	—	131k	13.4	13.4	13.4	In: $0.66 Out: $1.00
318	Qwen3 VL 30B A3B (Reasoning)Open Alibaba	13.3	110 t/s	2ms	256k	13.3	13.3	13.3	In: $0.20 Out: $2.40
319	Gemini 2.0 Flash Thinking Experimental (Jan '25) Google	13.3	—	—	1M	13.3	13.3	13.3	In: — Out: —
320	Gemma 4 12B (Non-reasoning)Open Google	13.2	108 t/s	2ms	262k	13.2	13.2	13.2	In: $0.10 Out: $0.30
321	Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning) Google	13.1	—	—	1M	13.1	13.1	13.1	In: $0.10 Out: $0.40
322	M Motif-2-12.7B-Reasoning Motif Technologies	12.8	—	—	128k	12.8	12.8	12.8	In: — Out: —
323	I Ling-1TOpen InclusionAI	12.8	—	—	128k	12.8	12.8	12.8	In: — Out: —
324	A Nova Premier Amazon	12.7	32 t/s	3ms	1M	12.7	12.7	12.7	In: $2.50 Out: $12.50
325	U Solar Pro 2 (Preview) (Reasoning) Upstage	12.5	—	—	64k	12.5	12.5	12.5	In: — Out: —
326	Magistral Medium 1 Mistral	12.5	—	—	40k	12.5	12.5	12.5	In: — Out: —
327	Mistral Medium 3 Mistral	12.5	48 t/s	1ms	128k	12.5	12.5	12.5	In: $0.40 Out: $2.00
328	M K2-V2 (medium)Open MBZUAI Institute of Foundation Models	12.4	—	—	512k	12.4	12.4	12.4	In: — Out: —
329	Llama Nemotron Super 49B v1.5 (Reasoning)Open NVIDIA	12.4	84 t/s	5ms	128k	12.4	12.4	12.4	In: $0.40 Out: $0.40
330	Devstral Medium Mistral	12.4	—	—	256k	12.4	12.4	12.4	In: — Out: —
331	Mistral Small 4 (Non-reasoning)Open Mistral	12.4	147 t/s	1ms	256k	12.4	12.4	12.4	In: $0.15 Out: $0.60
332	T Tri-21B-ThinkOpen Trillion Labs	12.4	—	—	32k	12.4	12.4	12.4	In: — Out: —
333	GPT-4o (March 2025, chatgpt-4o-latest) OpenAI	12.3	—	—	128k	12.3	12.3	12.3	In: — Out: —
334	Gemini 2.0 Flash (Feb '25) Google	12.3	—	—	1M	12.3	12.3	12.3	In: — Out: —
335	Claude 3.5 Haiku Anthropic	12.3	—	—	200k	12.3	12.3	12.3	In: — Out: —
336	Llama 3.3 Nemotron Super 49B v1 (Reasoning)Open NVIDIA	12.2	—	—	128k	12.2	12.2	12.2	In: — Out: —
337	Gemma 4 E4B (Reasoning)Open Google	12.2	85 t/s	1ms	128k	12.2	12.2	12.2	In: $0.02 Out: $0.10
338	Qwen3 4B 2507 (Reasoning)Open Alibaba	12.0	—	—	262k	12.0	12.0	12.0	In: — Out: —
339	O MiniCPM5-1B (Reasoning)Open OpenBMB	12.0	—	—	128k	12.0	12.0	12.0	In: — Out: —
340	S Sarvam 105B (high)Open Sarvam	11.9	—	—	128k	11.9	11.9	11.9	In: $0.04 Out: $0.17
341	Gemini 2.0 Pro Experimental (Feb '25) Google	11.8	—	—	2M	11.8	11.8	11.8	In: — Out: —
342	A Nova 2.0 Lite (Non-reasoning) Amazon	11.8	150 t/s	1ms	1M	11.8	11.8	11.8	In: $0.30 Out: $2.50
343	Devstral Small (May '25)Open Mistral	11.8	—	—	256k	11.8	11.8	11.8	In: — Out: —
344	C Celeris-1 Celeris	11.8	2158 t/s	1ms	131k	11.8	11.8	11.8	In: $0.20 Out: $0.70
345	Claude 3 Opus Anthropic	11.8	—	—	200k	11.8	11.8	11.8	In: $15.00 Out: $75.00
346	O MiniCPM5-1B (Non-reasoning)Open OpenBMB	11.7	—	—	128k	11.7	11.7	11.7	In: — Out: —
347	P Sonar Reasoning Perplexity	11.7	—	—	127k	11.7	11.7	11.7	In: — Out: —
348	Gemini 2.5 Flash Preview (Non-reasoning) Google	11.7	—	—	1M	11.7	11.7	11.7	In: — Out: —
349	Qwen3 32B (Reasoning)Open Alibaba	11.5	94 t/s	2ms	33k	11.5	11.5	11.5	In: $0.70 Out: $8.40
350	Gemini 2.5 Flash-Lite (Reasoning) Google	11.4	358 t/s	26ms	1M	11.4	11.4	11.4	In: $0.10 Out: $0.40
351	Magistral Small 1.2Open Mistral	11.3	147 t/s	1ms	128k	11.3	11.3	11.3	In: $0.50 Out: $1.50
352	GPT-4o (Nov '24) OpenAI	11.2	162 t/s	1ms	128k	11.2	11.2	11.2	In: $2.50 Out: $10.00
353	N Nanbeige4.1-3BOpen Nanbeige	11.1	—	—	256k	11.1	11.1	11.1	In: — Out: —
354	Qwen3 VL 32B InstructOpen Alibaba	11.1	76 t/s	3ms	256k	11.1	11.1	11.1	In: $0.70 Out: $2.80
355	Ministral 3 14BOpen Mistral	11.1	97 t/s	1ms	256k	11.1	11.1	11.1	In: $0.20 Out: $0.20
356	DeepSeek R1 Distill Qwen 32BOpen DeepSeek	11.0	—	—	128k	11.0	11.0	11.0	In: — Out: —
357	Z GLM-4.6V (Non-reasoning)Open Z AI	11.0	85 t/s	4ms	128k	11.0	11.0	11.0	In: $0.30 Out: $0.90
358	Qwen3 235B A22B (Non-reasoning)Open Alibaba	10.9	61 t/s	3ms	33k	10.9	10.9	10.9	In: $0.70 Out: $2.80
359	Magistral Small 1Open Mistral	10.7	—	—	40k	10.7	10.7	10.7	In: — Out: —
360	Gemini 2.0 Flash (experimental) Google	10.7	—	—	1M	10.7	10.7	10.7	In: $0.00 Out: $0.00
361	L EXAONE 4.0 32B (Reasoning)Open LG AI Research	10.6	—	—	131k	10.6	10.6	10.6	In: — Out: —
362	Qwen3 VL 8B (Reasoning)Open Alibaba	10.6	113 t/s	2ms	256k	10.6	10.6	10.6	In: $0.18 Out: $2.10
363	Mistral Small 3.2Open Mistral	10.6	138 t/s	1ms	128k	10.6	10.6	10.6	In: $0.10 Out: $0.30
364	A Nova 2.0 Omni (Non-reasoning) Amazon	10.5	—	—	1M	10.5	10.5	10.5	In: $0.30 Out: $2.50
365	Qwen3 14B (Reasoning)Open Alibaba	10.4	45 t/s	15ms	33k	10.4	10.4	10.4	In: $0.35 Out: $4.20
366	DeepSeek R1 0528 Qwen3 8BOpen DeepSeek	10.4	—	—	33k	10.4	10.4	10.4	In: — Out: —
367	Qwen2.5 Max Alibaba	10.2	—	—	32k	10.2	10.2	10.2	In: — Out: —
368	Llama 4 ScoutOpen Meta	10.0	131 t/s	1ms	10M	10.0	10.0	10.0	In: $0.18 Out: $0.66
369	Qwen3 VL 30B A3B InstructOpen Alibaba	10.0	114 t/s	2ms	256k	10.0	10.0	10.0	In: $0.20 Out: $0.80
370	N Hermes 4 - Llama-3.1 70B (Reasoning)Open Nous Research	10.0	90 t/s	1ms	128k	10.0	10.0	10.0	In: $0.13 Out: $0.40
371	Gemini 1.5 Pro (Sep '24) Google	10.0	—	—	2M	10.0	10.0	10.0	In: — Out: —
372	U Solar Pro 2 (Preview) (Non-reasoning) Upstage	10.0	—	—	64k	10.0	10.0	10.0	In: — Out: —
373	DeepSeek R1 Distill Llama 70BOpen DeepSeek	9.9	28 t/s	1ms	128k	9.9	9.9	9.9	In: $0.70 Out: $1.05
374	Claude 3.5 Sonnet (Oct '24) Anthropic	9.9	—	—	200k	9.9	9.9	9.9	In: $3.00 Out: $15.00
375	DeepSeek R1 Distill Qwen 14BOpen DeepSeek	9.8	—	—	128k	9.8	9.8	9.8	In: — Out: —
376	T Falcon-H1R-7BOpen TII UAE	9.8	—	—	256k	9.8	9.8	9.8	In: — Out: —
377	I Ling-flash-2.0Open InclusionAI	9.7	68 t/s	2ms	128k	9.7	9.7	9.7	In: $0.14 Out: $0.57
378	Gemma 4 E2B (Reasoning)Open Google	9.7	—	—	128k	9.7	9.7	9.7	In: — Out: —
379	Qwen3 Omni 30B A3B (Reasoning)Open Alibaba	9.6	100 t/s	2ms	66k	9.6	9.6	9.6	In: $0.25 Out: $0.97
380	GPT-4o (Aug '24) OpenAI	9.6	92 t/s	1ms	128k	9.6	9.6	9.6	In: $2.50 Out: $10.00
381	GPT-4.1 nano OpenAI	9.6	192 t/s	1ms	1M	9.6	9.6	9.6	In: $0.10 Out: $0.40
382	Qwen2.5 Instruct 72BOpen Alibaba	9.6	—	—	131k	9.6	9.6	9.6	In: $0.47 Out: $0.49
383	P Sonar Perplexity	9.5	—	—	127k	9.5	9.5	9.5	In: — Out: —
384	S Step3 VL 10BOpen StepFun	9.5	—	—	66k	9.5	9.5	9.5	In: — Out: —
385	Llama 3.3 Instruct 70BOpen Meta	9.4	84 t/s	1ms	128k	9.4	9.4	9.4	In: $0.60 Out: $0.72
386	Qwen3 30B A3B (Reasoning)Open Alibaba	9.3	104 t/s	2ms	33k	9.3	9.3	9.3	In: $0.20 Out: $2.40
387	P Sonar Pro Perplexity	9.3	—	—	200k	9.3	9.3	9.3	In: — Out: —
388	Devstral Small (Jul '25)Open Mistral	9.3	—	—	256k	9.3	9.3	9.3	In: — Out: —
389	QwQ 32B-PreviewOpen Alibaba	9.2	—	—	33k	9.2	9.2	9.2	In: — Out: —
390	Z GLM-4.5V (Reasoning)Open Z AI	9.1	82 t/s	2ms	64k	9.1	9.1	9.1	In: $0.60 Out: $1.80
391	Mistral Large 2 (Nov '24)Open Mistral	9.1	—	—	128k	9.1	9.1	9.1	In: — Out: —
392	Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)Open NVIDIA	9.1	52 t/s	2ms	128k	9.1	9.1	9.1	In: $0.60 Out: $1.80
393	Qwen3 30B A3B 2507 InstructOpen Alibaba	9.1	142 t/s	2ms	262k	9.1	9.1	9.1	In: $0.20 Out: $0.80
394	ERNIE 4.5 300B A47BOpen Baidu	9.0	—	—	131k	9.0	9.0	9.0	In: $0.28 Out: $1.10
395	N Hermes 4 - Llama-3.1 405B (Reasoning)Open Nous Research	9.0	34 t/s	3ms	128k	9.0	9.0	9.0	In: $1.00 Out: $3.00
396	U Solar Pro 2 (Reasoning) Upstage	9.0	—	—	66k	9.0	9.0	9.0	In: — Out: —
397	NVIDIA Nemotron Nano 12B v2 VL (Reasoning)Open NVIDIA	9.0	95 t/s	6ms	128k	9.0	9.0	9.0	In: $0.20 Out: $0.60
398	Ministral 3 8BOpen Mistral	9.0	118 t/s	1ms	256k	9.0	9.0	9.0	In: $0.15 Out: $0.15
399	Gemma 4 E4B (Non-reasoning)Open Google	8.9	87 t/s	1ms	128k	8.9	8.9	8.9	In: $0.02 Out: $0.10
400	I Granite 4.1 30BOpen IBM	8.9	—	—	131k	8.9	8.9	8.9	In: — Out: —
401	NVIDIA Nemotron Nano 9B V2 (Reasoning)Open NVIDIA	8.8	192 t/s	4ms	131k	8.8	8.8	8.8	In: $0.04 Out: $0.16
402	N Hermes 4 - Llama-3.1 405B (Non-reasoning)Open Nous Research	8.8	33 t/s	2ms	128k	8.8	8.8	8.8	In: $1.00 Out: $3.00
403	Gemini 2.0 Flash-Lite (Feb '25) Google	8.8	—	—	1M	8.8	8.8	8.8	In: — Out: —
404	NVIDIA Nemotron 3 Nano 4BOpen NVIDIA	8.7	—	—	262k	8.7	8.7	8.7	In: — Out: —
405	Llama Nemotron Super 49B v1.5 (Non-reasoning)Open NVIDIA	8.7	91 t/s	5ms	128k	8.7	8.7	8.7	In: $0.40 Out: $0.40
406	Qwen3 32B (Non-reasoning)Open Alibaba	8.6	97 t/s	2ms	33k	8.6	8.6	8.6	In: $0.70 Out: $2.80
407	GPT-4o (May '24) OpenAI	8.6	101 t/s	1ms	128k	8.6	8.6	8.6	In: $5.00 Out: $15.00
408	Gemini 2.0 Flash-Lite (Preview) Google	8.6	—	—	1M	8.6	8.6	8.6	In: — Out: —
409	M K2-V2 (low)Open MBZUAI Institute of Foundation Models	8.6	—	—	512k	8.6	8.6	8.6	In: — Out: —
410	Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)Open NVIDIA	8.5	—	—	128k	8.5	8.5	8.5	In: — Out: —
411	Kimi Linear 48B A3B InstructOpen Kimi	8.5	—	—	1M	8.5	8.5	8.5	In: — Out: —
412	Llama 3.1 Instruct 405BOpen Meta	8.5	—	—	128k	8.5	8.5	8.5	In: $2.50 Out: $10.00
413	Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)Open NVIDIA	8.5	—	—	128k	8.5	8.5	8.5	In: — Out: —
414	Qwen3 VL 8B InstructOpen Alibaba	8.4	119 t/s	2ms	256k	8.4	8.4	8.4	In: $0.18 Out: $0.70
415	Qwen3 4B (Reasoning)Open Alibaba	8.4	—	—	32k	8.4	8.4	8.4	In: — Out: —
416	L LFM2.5-8B-A1BOpen Liquid AI	8.3	337 t/s	2ms	33k	8.3	8.3	8.3	In: $0.00 Out: $0.00
417	Claude 3.5 Sonnet (June '24) Anthropic	8.3	—	—	200k	8.3	8.3	8.3	In: $3.00 Out: $15.00
418	Qwen3 8B (Reasoning)Open Alibaba	8.3	39 t/s	4ms	131k	8.3	8.3	8.3	In: $0.18 Out: $2.10
419	A Llama 3.1 Tulu3 405BOpen Allen Institute for AI	8.3	—	—	128k	8.3	8.3	8.3	In: — Out: —
420	GPT-4o (ChatGPT) OpenAI	8.2	—	—	128k	8.2	8.2	8.2	In: — Out: —
421	I Ring-flash-2.0Open InclusionAI	8.2	—	—	128k	8.2	8.2	8.2	In: $0.14 Out: $0.57
422	Pixtral LargeOpen Mistral	8.1	—	—	128k	8.1	8.1	8.1	In: — Out: —
423	A Olmo 3.1 32B ThinkOpen Allen Institute for AI	8.1	—	—	66k	8.1	8.1	8.1	In: $0.00 Out: $0.00
424	Grok 2 (Dec '24)Open SpaceXAI	8.0	—	—	131k	8.0	8.0	8.0	In: — Out: —
425	GPT-5 nano (minimal) OpenAI	8.0	144 t/s	1ms	400k	8.0	8.0	8.0	In: $0.05 Out: $0.40
426	Gemini 1.5 Flash (Sep '24) Google	8.0	—	—	1M	8.0	8.0	8.0	In: — Out: —
427	Qwen3 VL 4B (Reasoning)Open Alibaba	7.9	—	—	256k	7.9	7.9	7.9	In: — Out: —
428	GPT-4 Turbo OpenAI	7.9	32 t/s	3ms	128k	7.9	7.9	7.9	In: $10.00 Out: $30.00
429	U Solar Pro 2 (Non-reasoning) Upstage	7.8	—	—	66k	7.8	7.8	7.8	In: — Out: —
430	A Nova Pro Amazon	7.7	—	—	300k	7.7	7.7	7.7	In: $0.80 Out: $3.20
431	C Command AOpen Cohere	7.7	59 t/s	2ms	256k	7.7	7.7	7.7	In: $2.50 Out: $10.00
432	Qwen3.5 2B (Reasoning)Open Alibaba	7.6	—	—	262k	7.6	7.6	7.6	In: — Out: —
433	Llama 3.1 Nemotron Instruct 70BOpen NVIDIA	7.6	96 t/s	7ms	128k	7.6	7.6	7.6	In: $1.20 Out: $1.20
434	Llama 3.1 Instruct 8BOpen Meta	7.6	141 t/s	1ms	128k	7.6	7.6	7.6	In: $0.08 Out: $0.09
435	Grok Beta SpaceXAI	7.5	—	—	128k	7.5	7.5	7.5	In: — Out: —
436	Qwen2.5 Instruct 32BOpen Alibaba	7.5	—	—	128k	7.5	7.5	7.5	In: — Out: —
437	Gemma 3 27B InstructOpen Google	7.4	—	—	128k	7.4	7.4	7.4	In: $0.00 Out: $0.00
438	NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)Open NVIDIA	7.4	205 t/s	1ms	1M	7.4	7.4	7.4	In: $0.05 Out: $0.20
439	NVIDIA Nemotron Nano 9B V2 (Non-reasoning)Open NVIDIA	7.4	172 t/s	2ms	131k	7.4	7.4	7.4	In: $0.05 Out: $0.20
440	Mistral Large 2 (Jul '24)Open Mistral	7.3	—	—	128k	7.3	7.3	7.3	In: $2.00 Out: $6.00
441	Qwen3 4B 2507 InstructOpen Alibaba	7.1	—	—	262k	7.1	7.1	7.1	In: — Out: —
442	Qwen2.5 Coder Instruct 32BOpen Alibaba	7.1	—	—	131k	7.1	7.1	7.1	In: — Out: —
443	Qwen3 14B (Non-reasoning)Open Alibaba	7.0	56 t/s	5ms	33k	7.0	7.0	7.0	In: $0.35 Out: $1.40
444	GPT-4 OpenAI	7.0	—	—	8k	7.0	7.0	7.0	In: $30.00 Out: $60.00
445	Z GLM-4.5V (Non-reasoning)Open Z AI	7.0	81 t/s	2ms	64k	7.0	7.0	7.0	In: $0.60 Out: $1.80
446	Mistral Small 3Open Mistral	6.9	146 t/s	1ms	32k	6.9	6.9	6.9	In: $0.10 Out: $0.30
447	Gemini 2.5 Flash-Lite (Non-reasoning) Google	6.9	232 t/s	0ms	1M	6.9	6.9	6.9	In: $0.10 Out: $0.40
448	A Nova Lite Amazon	6.9	162 t/s	1ms	300k	6.9	6.9	6.9	In: $0.06 Out: $0.24
449	GPT-4o mini OpenAI	6.9	82 t/s	1ms	128k	6.9	6.9	6.9	In: $0.15 Out: $0.60
450	N Hermes 4 - Llama-3.1 70B (Non-reasoning)Open Nous Research	6.9	95 t/s	1ms	128k	6.9	6.9	6.9	In: $0.13 Out: $0.40
451	Qwen3 30B A3B (Non-reasoning)Open Alibaba	6.8	108 t/s	2ms	33k	6.8	6.8	6.8	In: $0.20 Out: $0.80
452	Ministral 3 3BOpen Mistral	6.8	203 t/s	1ms	256k	6.8	6.8	6.8	In: $0.10 Out: $0.10
453	DeepSeek-V2.5 (Dec '24)Open DeepSeek	6.8	—	—	128k	6.8	6.8	6.8	In: — Out: —
454	Qwen3 4B (Non-reasoning)Open Alibaba	6.8	—	—	32k	6.8	6.8	6.8	In: — Out: —
455	Llama 3.1 Instruct 70BOpen Meta	6.8	61 t/s	1ms	128k	6.8	6.8	6.8	In: $0.56 Out: $0.56
456	I Granite 4.1 8BOpen IBM	6.7	111 t/s	1ms	131k	6.7	6.7	6.7	In: $0.05 Out: $0.10
457	S Sarvam 30B (high)Open Sarvam	6.6	—	—	66k	6.6	6.6	6.6	In: $0.03 Out: $0.11
458	Gemini 2.0 Flash Thinking Experimental (Dec '24) Google	6.6	—	—	2M	6.6	6.6	6.6	In: — Out: —
459	DeepSeek-V2.5Open DeepSeek	6.6	—	—	128k	6.6	6.6	6.6	In: — Out: —
460	A Olmo 3.1 32B InstructOpen Allen Institute for AI	6.5	—	—	66k	6.5	6.5	6.5	In: — Out: —
461	Mistral Saba Mistral	6.4	—	—	32k	6.4	6.4	6.4	In: — Out: —
462	DeepSeek R1 Distill Llama 8BOpen DeepSeek	6.4	—	—	128k	6.4	6.4	6.4	In: — Out: —
463	Gemma 4 E2B (Non-reasoning)Open Google	6.4	—	—	128k	6.4	6.4	6.4	In: — Out: —
464	A Olmo 3 32B ThinkOpen Allen Institute for AI	6.4	—	—	66k	6.4	6.4	6.4	In: — Out: —
465	Gemini 1.5 Pro (May '24) Google	6.3	—	—	2M	6.3	6.3	6.3	In: — Out: —
466	P R1 1776Open Perplexity	6.3	—	—	128k	6.3	6.3	6.3	In: — Out: —
467	Qwen2.5 Turbo Alibaba	6.3	107 t/s	2ms	1M	6.3	6.3	6.3	In: $0.05 Out: $0.20
468	R Reka Flash (Sep '24) Reka AI	6.3	57 t/s	2ms	128k	6.3	6.3	6.3	In: $0.20 Out: $0.80
469	Llama 3.2 Instruct 90B (Vision)Open Meta	6.2	—	—	128k	6.2	6.2	6.2	In: — Out: —
470	U Solar MiniOpen Upstage	6.2	—	—	4k	6.2	6.2	6.2	In: $0.15 Out: $0.15
471	Grok-1Open SpaceXAI	6.0	—	—	8k	6.0	6.0	6.0	In: — Out: —
472	Qwen2 Instruct 72BOpen Alibaba	6.0	—	—	131k	6.0	6.0	6.0	In: — Out: —
473	Phi-4 Mini InstructOpen Microsoft	6.0	44 t/s	1ms	128k	6.0	6.0	6.0	In: $0.00 Out: $0.00
474	L EXAONE 4.0 32B (Non-reasoning)Open LG AI Research	6.0	—	—	131k	6.0	6.0	6.0	In: — Out: —
475	Qwen3.5 2B (Non-reasoning)Open Alibaba	5.6	—	—	262k	5.6	5.6	5.6	In: — Out: —
476	Gemini 1.5 Flash-8B Google	5.5	—	—	1M	5.5	5.5	5.5	In: — Out: —
477	Gemma 3 12B InstructOpen Google	5.5	—	—	128k	5.5	5.5	5.5	In: $0.00 Out: $0.00
478	Qwen3.5 0.8B (Reasoning)Open Alibaba	5.5	—	—	262k	5.5	5.5	5.5	In: — Out: —
479	N DeepHermes 3 - Mistral 24B Preview (Non-reasoning)Open Nous Research	5.3	—	—	32k	5.3	5.3	5.3	In: — Out: —
480	A Jamba 1.7 LargeOpen AI21 Labs	5.3	58 t/s	1ms	256k	5.3	5.3	5.3	In: $2.00 Out: $8.00
481	I Granite 4.0 H SmallOpen IBM	5.2	48 t/s	10ms	128k	5.2	5.2	5.2	In: $0.06 Out: $0.25
482	A Jamba 1.5 LargeOpen AI21 Labs	5.1	—	—	256k	5.1	5.1	5.1	In: $2.00 Out: $8.00
483	Qwen3 Omni 30B A3B InstructOpen Alibaba	5.1	95 t/s	2ms	66k	5.1	5.1	5.1	In: $0.25 Out: $0.97
484	N Hermes 3 - Llama-3.1 70BOpen Nous Research	5.1	33 t/s	2ms	128k	5.1	5.1	5.1	In: $0.70 Out: $0.70
485	Qwen3 8B (Non-reasoning)Open Alibaba	5.1	39 t/s	4ms	33k	5.1	5.1	5.1	In: $0.18 Out: $0.70
486	DeepSeek-Coder-V2Open DeepSeek	5.1	—	—	128k	5.1	5.1	5.1	In: — Out: —
487	A OLMo 2 32BOpen Allen Institute for AI	5.0	—	—	4k	5.0	5.0	5.0	In: — Out: —
488	A Jamba 1.6 LargeOpen AI21 Labs	5.0	56 t/s	1ms	256k	5.0	5.0	5.0	In: $2.00 Out: $8.00
489	L LFM2 24B A2BOpen Liquid AI	4.9	—	—	33k	4.9	4.9	4.9	In: — Out: —
490	Gemini 1.5 Flash (May '24) Google	4.9	—	—	1M	4.9	4.9	4.9	In: — Out: —
491	Phi-4Open Microsoft	4.9	—	—	16k	4.9	4.9	4.9	In: $0.13 Out: $0.50
492	Claude 3 Sonnet Anthropic	4.7	—	—	200k	4.7	4.7	4.7	In: $3.00 Out: $15.00
493	A Nova Micro Amazon	4.7	279 t/s	1ms	130k	4.7	4.7	4.7	In: $0.04 Out: $0.14
494	I Granite 4.1 3BOpen IBM	4.7	—	—	131k	4.7	4.7	4.7	In: — Out: —
495	Mistral Small (Sep '24)Open Mistral	4.7	147 t/s	1ms	33k	4.7	4.7	4.7	In: $0.20 Out: $0.60
496	Gemini 1.0 Ultra Google	4.6	—	—	33k	4.6	4.6	4.6	In: — Out: —
497	Phi-3 Mini Instruct 3.8BOpen Microsoft	4.6	—	—	4k	4.6	4.6	4.6	In: — Out: —
498	NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)Open NVIDIA	4.6	166 t/s	2ms	128k	4.6	4.6	4.6	In: $0.20 Out: $0.60
499	Gemma 3n E4B Instruct Preview (May '25)Open Google	4.6	—	—	32k	4.6	4.6	4.6	In: — Out: —
500	Phi-4 Multimodal InstructOpen Microsoft	4.5	18 t/s	1ms	128k	4.5	4.5	4.5	In: $0.00 Out: $0.00
501	Qwen2.5 Coder Instruct 7B Open Alibaba	4.5	—	—	131k	4.5	4.5	4.5	In: — Out: —
502	Mistral Large (Feb '24) Mistral	4.4	—	—	33k	4.4	4.4	4.4	In: $4.00 Out: $12.00
503	Mixtral 8x22B InstructOpen Mistral	4.4	—	—	65k	4.4	4.4	4.4	In: — Out: —
504	Llama 2 Chat 7BOpen Meta	4.3	—	—	4k	4.3	4.3	4.3	In: $0.05 Out: $0.25
505	Llama 3.2 Instruct 3BOpen Meta	4.2	—	—	128k	4.2	4.2	4.2	In: — Out: —
506	O MiniCPM-V 4.6 1.3BOpen OpenBMB	4.2	—	—	262k	4.2	4.2	4.2	In: — Out: —
507	A Jamba Reasoning 3BOpen AI21 Labs	4.1	—	—	262k	4.1	4.1	4.1	In: — Out: —
508	Qwen3 VL 4B InstructOpen Alibaba	4.1	—	—	256k	4.1	4.1	4.1	In: — Out: —
509	Qwen1.5 Chat 110BOpen Alibaba	4.1	—	—	32k	4.1	4.1	4.1	In: — Out: —
510	R Reka Flash 3Open Reka AI	4.1	—	—	128k	4.1	4.1	4.1	In: $0.20 Out: $0.80
511	A Olmo 3 7B ThinkOpen Allen Institute for AI	4.0	—	—	66k	4.0	4.0	4.0	In: — Out: —
512	Claude 2.1 Anthropic	3.9	—	—	200k	3.9	3.9	3.9	In: — Out: —
513	Claude 3 Haiku Anthropic	3.9	—	—	200k	3.9	3.9	3.9	In: $0.25 Out: $1.25
514	A OLMo 2 7BOpen Allen Institute for AI	3.9	—	—	4k	3.9	3.9	3.9	In: — Out: —
515	A Molmo 7B-DOpen Allen Institute for AI	3.8	—	—	4k	3.8	3.8	3.8	In: — Out: —
516	I Ling-mini-2.0Open InclusionAI	3.8	—	—	131k	3.8	3.8	3.8	In: — Out: —
517	DeepSeek R1 Distill Qwen 1.5BOpen DeepSeek	3.7	—	—	128k	3.7	3.7	3.7	In: — Out: —
518	DeepSeek-V2-ChatOpen DeepSeek	3.6	—	—	128k	3.6	3.6	3.6	In: — Out: —
519	Claude 2.0 Anthropic	3.6	—	—	100k	3.6	3.6	3.6	In: — Out: —
520	Mistral Small (Feb '24) Mistral	3.6	145 t/s	1ms	33k	3.6	3.6	3.6	In: $0.15 Out: $0.60
521	Mistral Medium Mistral	3.6	133 t/s	2ms	33k	3.6	3.6	3.6	In: $1.50 Out: $7.50
522	GPT-3.5 Turbo OpenAI	3.6	—	—	4k	3.6	3.6	3.6	In: $0.50 Out: $1.50
523	Llama 3 Instruct 70BOpen Meta	3.5	—	—	8k	3.5	3.5	3.5	In: $0.65 Out: $2.75
524	Qwen Chat 72BOpen Alibaba	3.4	—	—	34k	3.4	3.4	3.4	In: — Out: —
525	S Arctic InstructOpen Snowflake	3.4	—	—	4k	3.4	3.4	3.4	In: — Out: —
526	L LFM 40B Liquid AI	3.4	—	—	32k	3.4	3.4	3.4	In: — Out: —
527	Llama 3.2 Instruct 11B (Vision)Open Meta	3.3	49 t/s	1ms	128k	3.3	3.3	3.3	In: $0.34 Out: $0.34
528	Qwen3.5 0.8B (Non-reasoning)Open Alibaba	3.3	—	—	262k	3.3	3.3	3.3	In: — Out: —
529	PALM-2 Google	3.2	—	—	8k	3.2	3.2	3.2	In: — Out: —
530	Gemini 1.0 Pro Google	3.1	—	—	33k	3.1	3.1	3.1	In: — Out: —
531	DeepSeek Coder V2 Lite InstructOpen DeepSeek	3.1	—	—	128k	3.1	3.1	3.1	In: — Out: —
532	S Sarvam M (Reasoning)Open Sarvam	3.0	—	—	33k	3.0	3.0	3.0	In: $0.00 Out: $0.00
533	Llama 2 Chat 70BOpen Meta	3.0	—	—	4k	3.0	3.0	3.0	In: — Out: —
534	DeepSeek LLM 67B Chat (V1)Open DeepSeek	3.0	—	—	4k	3.0	3.0	3.0	In: — Out: —
535	Llama 2 Chat 13BOpen Meta	3.0	—	—	4k	3.0	3.0	3.0	In: — Out: —
536	C Command-R+ (Apr '24)Open Cohere	3.0	—	—	128k	3.0	3.0	3.0	In: $3.00 Out: $15.00
537	O OpenChat 3.5 (1210)Open OpenChat	3.0	—	—	8k	3.0	3.0	3.0	In: — Out: —
538	D DBRX InstructOpen Databricks	3.0	—	—	33k	3.0	3.0	3.0	In: — Out: —
539	L Exaone 4.0 1.2B (Reasoning)Open LG AI Research	2.9	—	—	64k	2.9	2.9	2.9	In: — Out: —
540	A Olmo 3 7B InstructOpen Allen Institute for AI	2.8	—	—	66k	2.8	2.8	2.8	In: $0.10 Out: $0.20
541	L Exaone 4.0 1.2B (Non-reasoning)Open LG AI Research	2.8	—	—	64k	2.8	2.8	2.8	In: — Out: —
542	L LFM2.5-1.2B-ThinkingOpen Liquid AI	2.7	—	—	32k	2.7	2.7	2.7	In: — Out: —
543	A Jamba 1.7 MiniOpen AI21 Labs	2.7	—	—	258k	2.7	2.7	2.7	In: — Out: —
544	L LFM2 2.6BOpen Liquid AI	2.7	—	—	33k	2.7	2.7	2.7	In: — Out: —
545	L LFM2.5-1.2B-InstructOpen Liquid AI	2.7	—	—	32k	2.7	2.7	2.7	In: — Out: —
546	A Jamba 1.5 MiniOpen AI21 Labs	2.7	—	—	256k	2.7	2.7	2.7	In: $0.20 Out: $0.40
547	I Granite 4.0 H 1BOpen IBM	2.7	—	—	128k	2.7	2.7	2.7	In: — Out: —
548	Qwen3 1.7B (Reasoning)Open Alibaba	2.6	—	—	32k	2.6	2.6	2.6	In: — Out: —
549	A Jamba 1.6 MiniOpen AI21 Labs	2.6	178 t/s	1ms	256k	2.6	2.6	2.6	In: $0.20 Out: $0.40
550	Mixtral 8x7B InstructOpen Mistral	2.4	—	—	33k	2.4	2.4	2.4	In: $0.45 Out: $0.70
551	Gemma 3 270MOpen Google	2.4	—	—	32k	2.4	2.4	2.4	In: — Out: —
552	S Apertus 70B InstructOpen Swiss AI Initiative	2.4	—	—	66k	2.4	2.4	2.4	In: $0.82 Out: $2.92
553	I Granite 4.0 MicroOpen IBM	2.4	—	—	128k	2.4	2.4	2.4	In: — Out: —
554	N DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)Open Nous Research	2.3	—	—	128k	2.3	2.3	2.3	In: — Out: —
555	Mistral 7B InstructOpen Mistral	2.1	120 t/s	1ms	8k	2.1	2.1	2.1	In: $0.25 Out: $0.25
556	Claude Instant Anthropic	2.1	—	—	100k	2.1	2.1	2.1	In: — Out: —
557	C Command-R (Mar '24)Open Cohere	2.1	—	—	128k	2.1	2.1	2.1	In: $0.50 Out: $1.50
558	Qwen Chat 14BOpen Alibaba	2.1	—	—	8k	2.1	2.1	2.1	In: — Out: —
559	Llama 65BOpen Meta	2.1	—	—	2k	2.1	2.1	2.1	In: — Out: —
560	I Granite 4.0 1BOpen IBM	2.1	—	—	128k	2.1	2.1	2.1	In: — Out: —
561	A Molmo2-8BOpen Allen Institute for AI	2.0	—	—	37k	2.0	2.0	2.0	In: — Out: —
562	L LFM2 8B A1BOpen Liquid AI	1.8	—	—	33k	1.8	1.8	1.8	In: — Out: —
563	I Granite 3.3 8B (Non-reasoning)Open IBM	1.8	15 t/s	27ms	128k	1.8	1.8	1.8	In: $0.03 Out: $0.25
564	Qwen3 1.7B (Non-reasoning)Open Alibaba	1.5	—	—	32k	1.5	1.5	1.5	In: — Out: —
565	Qwen3 0.6B (Reasoning)Open Alibaba	1.3	—	—	32k	1.3	1.3	1.3	In: — Out: —
566	Llama 3 Instruct 8BOpen Meta	1.2	—	—	8k	1.2	1.2	1.2	In: $0.04 Out: $0.14
567	Gemma 3n E4B InstructOpen Google	1.2	43 t/s	1ms	32k	1.2	1.2	1.2	In: $0.06 Out: $0.12
568	L LFM2 1.2BOpen Liquid AI	1.1	—	—	33k	1.1	1.1	1.1	In: — Out: —
569	Gemma 3 4B InstructOpen Google	1.1	—	—	128k	1.1	1.1	1.1	In: $0.00 Out: $0.00
570	Llama 3.2 Instruct 1BOpen Meta	1.1	—	—	128k	1.1	1.1	1.1	In: — Out: —
571	L LFM2.5-VL-1.6BOpen Liquid AI	1.0	398 t/s	1ms	32k	1.0	1.0	1.0	In: $0.00 Out: $0.00
572	Qwen3 0.6B (Non-reasoning)Open Alibaba	1.0	—	—	32k	1.0	1.0	1.0	In: — Out: —
573	I Granite 4.0 350MOpen IBM	1.0	—	—	33k	1.0	1.0	1.0	In: — Out: —
574	C Tiny Aya GlobalOpen Cohere	1.0	—	—	8k	1.0	1.0	1.0	In: $0.00 Out: $0.00
575	S Apertus 8B InstructOpen Swiss AI Initiative	1.0	—	—	66k	1.0	1.0	1.0	In: $0.10 Out: $0.20
576	I Granite 4.0 H 350MOpen IBM	1.0	—	—	33k	1.0	1.0	1.0	In: — Out: —
577	Gemma 3n E2B InstructOpen Google	1.0	—	—	32k	1.0	1.0	1.0	In: $0.00 Out: $0.00
578	Gemma 3 1B InstructOpen Google	1.0	—	—	32k	1.0	1.0	1.0	In: $0.00 Out: $0.00