LLM Leaderboard
Compare quality, price, and speed across today's leading models
Top Klu Index100Claude Fable 5.2
Best value$3Gemini 4 Hydrogen
Fastest output439 tok/sMuse Ember
Fastest first token0.7sNorth Mini Code
Largest context1.05M tokensGPT-6 Sol
Most expensive$20 / 1MClaude Fable 5.1
Klu Frontier IndexPrice per 1M tokens, log scale, cheaper to the right
FrontierAnthropicOpenAIGooglexAIMetaMistralOther
Frontier explorerKlu Index against cost per task, a point per reasoning effort
ExploreKlu Index
67 models
- All labs67
- Anthropic6
- Google7
- Meta6
- OpenAI7
- Alibaba3
- Thinking Machines3
- Mistral4
- NVIDIA4
- Other labs27
| # | Model | Klu Index | $ / 1M tokens | Tokens / s | First token | Context | Frontier |
|---|---|---|---|---|---|---|---|
| 1 | Claude Fable 5.2Anthropic | 100 | $16.00 | 101 | 271s | 1M | Frontier |
| 2 | Claude Opus 5.5Anthropic | 94 | $8.00 | 114 | 390s | 1M | Frontier |
| 3 | Gemini 4 CarbonGoogle | 93 | $4.00 | 66 | 71s | 1M | Frontier |
| 4 | Gemini 4 HydrogenGoogle | 91 | $3.00 | 181 | 45s | 1M | Frontier |
| 5 | Claude Sonnet 5.5Anthropic | 90 | $4.00 | 138 | 266s | 1M | No |
| 6 | Muse Wonder FullMuse Code subscription | 87 | Subscription | — | — | 1M | No |
| 7 | Claude Fable 5.1Anthropic | 84 | $20.00 | 88 | 204s | 1M | No |
| 8 | Kimi K3.1 ProKimi | 84 | $4.00 | 30 | 4.5s | 1M | No |
| 9 | Gemini 4 ArgonGoogle | 83 | $4.00 | 65 | 45s | 1M | No |
| 10 | Muse WonderMeta Model API and Muse Code | 83 | $8.00 | 110 | 26s | 1M | No |
| 11 | GPT-6 AstraOpenAI | 82 | $20.00 | 36 | 250s | 1M | No |
| 12 | GPT-6.1 SolOpenAI | 82 | $4.00 | 48 | 201s | 1M | No |
| 13 | Muse EmberMeta | 81 | $0.20 | 439 | 5.0s | 1M | Frontier |
| 14 | Claude Opus 5Anthropic | 80 | $10.00 | 81 | 55s | 1M | No |
| 15 | DeepSeek V4.1 Flash 1017DeepSeek | 79 | $0.53 | 162 | 1.3s | 1M | No |
| 16 | GPT-6 SolOpenAI | 76 | $4.00 | 111 | 83s | 1.05M | No |
| 17 | GPT-6.1 TerraOpenAI | 76 | $2.00 | 68 | 188s | 1M | No |
| 18 | Muse Spark 1.3Meta | 76 | $2.00 | 132 | 29s | 1M | No |
| 19 | Step 5 PreviewStepFun | 73 | $1.43 | 133 | 0.9s | 1M | No |
| 20 | Claude Haiku 5.5Anthropic | 72 | $0.20 | 252 | 175s | 1M | No |
| 21 | Grok 4.7xAI | 72 | $3.00 | 65 | 64s | 500k | No |
| 22 | Qwen3.8 Max (0902)Alibaba | 72 | $3.00 | 58 | 4.7s | 984k | No |
| 23 | Tencent Hy4Tencent | 72 | $1.25 | 91 | 3.0s | 1M | No |
| 24 | GLM-5.3Z AI | 71 | $2.15 | 75 | 3.1s | 1M | No |
| 25 | Inkling 2 SmallThinking Machines | 71 | $0.48 | 76 | 3.5s | 1.02M | No |
| 26 | MiniMax-M3.1-FlashMiniMax | 71 | $0.26 | 139 | 1.6s | 1M | No |
| 27 | Ling 3.1 Flash 1019InclusionAI | 69 | $0.45 | — | — | 1M | No |
| 28 | Grok 4.6xAI | 68 | $3.00 | 64 | 33s | 500k | No |
| 29 | Muse Spark OpenMeta | 68 | $2.00 | 146 | 1.2s | 1M | No |
| 30 | Qwen3.8-Flash-NextAlibaba | 68 | $0.23 | 68 | 2.6s | 256k | No |
| 31 | Agnes 2.5 Pro ThetaSapiens AI | 67 | $0.15 | 170 | 2.1s | 1M | Frontier |
| 32 | MiMo-V2.6-ProXiaomi | 67 | $0.54 | 56 | 4.7s | 1M | No |
| 33 | Gemini 3.8 FlashGoogle | 65 | $1.50 | 113 | 17s | 1M | No |
| 34 | GLM-5.3-FlashZ AI | 64 | $0.24 | 47 | 3.7s | 1M | No |
| 35 | Mistral Large 4Mistral API and open weights | 64 | $1.05 | 108 | 3.4s | 1M | No |
| 36 | MiMo-V2.6-FlashXiaomi | 62 | $0.18 | 43 | 5.3s | 1M | No |
| 37 | Mistral Large 4 PreviewRetired public preview | 62 | $1.05 | 98 | 6.3s | 512k | No |
| 38 | GPT-5.6 TerraOpenAI | 61 | $4.50 | 99 | 106s | 1M | No |
| 39 | Beam 1.0Reflection | 58 | $0.60 | 271 | 0.8s | 1M | No |
| 40 | DeepSeek V4 Pro 0813DeepSeek | 58 | $1.98 | 80 | 2.7s | 1M | No |
| 41 | GPT-6 LunaOpenAI | 58 | $0.20 | 105 | 89s | 1M | No |
| 42 | K2 Horizon 375B A23BInstitute of Foundation Models | 58 | — | 250 | 4.8s | 512k | No |
| 43 | GPT-5.6 LunaOpenAI | 55 | $0.45 | 106 | 76s | 1M | No |
| 44 | Nanbeige4.3-33BNanbeige | 54 | — | — | — | 512k | No |
| 45 | Qwen3.8 27BAlibaba | 54 | $1.13 | 41 | 5.7s | 256k | No |
| 46 | JT-4.1 Flash 236B A21BChina Mobile | 52 | — | — | — | 256k | No |
| 47 | MiniMax-M3MiniMax | 52 | $0.53 | 109 | 1.8s | 1M | No |
| 48 | Nemotron 3.5 Super VLNVIDIA | 51 | $0.26 | 221 | 1.1s | 1M | No |
| 49 | Apodex 1.1Apodex | 46 | $0.98 | 131 | 2.3s | 256k | No |
| 50 | Solar Pro 4Upstage | 46 | $0.53 | 84 | 2.3s | 512k | No |
| 51 | Nemotron 3 Ultra 550B A55BNVIDIA | 45 | $1.13 | 199 | 1.9s | 256k | No |
| 52 | Inkling SmallThinking Machines | 44 | $0.53 | 98 | 1.5s | 1M | No |
| 53 | Kimi K3Kimi | 44 | $6.00 | 42 | 7.4s | 1M | No |
| 54 | Gemini 3.5 Flash-LiteGoogle | 43 | $0.85 | 288 | 4.7s | 1M | No |
| 55 | InklingThinking Machines | 41 | $1.76 | 75 | 1.9s | 1M | No |
| 56 | K-EXAONE 2.0LG AI Research | 34 | — | — | — | 256k | No |
| 57 | Nemotron 3.5 LightningNVIDIA | 33 | $0.10 | 259 | 0.9s | 1M | Frontier |
| 58 | Command A+Cohere | 32 | — | 196 | 0.9s | 192k | No |
| 59 | Gemini 3.1 Flash-LiteGoogle | 30 | $0.56 | 305 | 4.1s | 1M | No |
| 60 | Granite 4.2 3BIBM | 27 | — | — | — | 128k | No |
| 61 | Muse GlimmerMeta | 27 | $0.53 | 77 | 1.8s | 128k | No |
| 62 | North Mini CodeCohere | 27 | — | 177 | 0.7s | 256k | No |
| 63 | Mistral Medium 3.5Mistral | 26 | $3.00 | 146 | 7.2s | 256k | No |
| 64 | Mistral Small 4Mistral | 26 | $0.26 | 144 | 1.5s | 256k | No |
| 65 | Granite 4.2 8BIBM | 22 | $0.11 | 80 | 6.6s | 128k | No |
| 66 | NVIDIA Nemotron 3 Nano 4BNVIDIA | 21 | — | — | — | 262k | No |
| 67 | Gemma 4 31BGoogle | 20 | — | 30 | 1.0s | 256k | No |
Muse Wonder Full is included in Muse Code Power Usage ($50/month) and Maximum Usage ($200/month). API token pricing and throughput are unavailable.
Retired public preview: The October 6 preview is retired. Historical preview measurements remain distinct from the final Mistral Large 4 release.
Compare
- Claude Fable 5.2Anthropic100
- Claude Opus 5.5Anthropic94
- Gemini 4 CarbonGoogle93
- Gemini 4 HydrogenGoogle91
- Claude Sonnet 5.5Anthropic90
- Muse Wonder FullMuse Code subscription87
- Claude Fable 5.1Anthropic84
- Kimi K3.1 ProKimi84
- Gemini 4 ArgonGoogle83
- Muse WonderMeta Model API and Muse Code83
- GPT-6 AstraOpenAI82
- GPT-6.1 SolOpenAI82
- Muse EmberMeta81
- Claude Opus 5Anthropic80
- DeepSeek V4.1 Flash 1017DeepSeek79
- GPT-6 SolOpenAI76
- GPT-6.1 TerraOpenAI76
- Muse Spark 1.3Meta76
- Step 5 PreviewStepFun73
- Claude Haiku 5.5Anthropic72
- Grok 4.7xAI72
- Qwen3.8 Max (0902)Alibaba72
- Tencent Hy4Tencent72
- GLM-5.3Z AI71
- Inkling 2 SmallThinking Machines71
- MiniMax-M3.1-FlashMiniMax71
- Ling 3.1 Flash 1019InclusionAI69
- Grok 4.6xAI68
- Muse Spark OpenMeta68
- Qwen3.8-Flash-NextAlibaba68
- Agnes 2.5 Pro ThetaSapiens AI67
- MiMo-V2.6-ProXiaomi67
- Gemini 3.8 FlashGoogle65
- GLM-5.3-FlashZ AI64
- Mistral Large 4Mistral API and open weights64
- MiMo-V2.6-FlashXiaomi62
- Mistral Large 4 PreviewRetired public preview62
- GPT-5.6 TerraOpenAI61
- Beam 1.0Reflection58
- DeepSeek V4 Pro 0813DeepSeek58
- GPT-6 LunaOpenAI58
- K2 Horizon 375B A23BInstitute of Foundation Models58
- GPT-5.6 LunaOpenAI55
- Nanbeige4.3-33BNanbeige54
- Qwen3.8 27BAlibaba54
- JT-4.1 Flash 236B A21BChina Mobile52
- MiniMax-M3MiniMax52
- Nemotron 3.5 Super VLNVIDIA51
- Apodex 1.1Apodex46
- Solar Pro 4Upstage46
- Nemotron 3 Ultra 550B A55BNVIDIA45
- Inkling SmallThinking Machines44
- Kimi K3Kimi44
- Gemini 3.5 Flash-LiteGoogle43
- InklingThinking Machines41
- K-EXAONE 2.0LG AI Research34
- Nemotron 3.5 LightningNVIDIA33
- Command A+Cohere32
- Gemini 3.1 Flash-LiteGoogle30
- Granite 4.2 3BIBM27
- Muse GlimmerMeta27
- North Mini CodeCohere27
- Mistral Medium 3.5Mistral26
- Mistral Small 4Mistral26
- Granite 4.2 8BIBM22
- NVIDIA Nemotron 3 Nano 4BNVIDIA21
- Gemma 4 31BGoogle20
- Claude Fable 5.2Anthropic100
- Claude Opus 5.5Anthropic94
- Gemini 4 CarbonGoogle93
- Gemini 4 HydrogenGoogle91
- Claude Sonnet 5.5Anthropic90
- Muse Wonder FullMuse Code subscription87
- Claude Fable 5.1Anthropic84
- Kimi K3.1 ProKimi84
- Gemini 4 ArgonGoogle83
- Muse WonderMeta Model API and Muse Code83
- GPT-6 AstraOpenAI82
- GPT-6.1 SolOpenAI82
- Muse EmberMeta81
- Claude Opus 5Anthropic80
- DeepSeek V4.1 Flash 1017DeepSeek79
- GPT-6 SolOpenAI76
- GPT-6.1 TerraOpenAI76
- Muse Spark 1.3Meta76
- Step 5 PreviewStepFun73
- Claude Haiku 5.5Anthropic72
- Grok 4.7xAI72
- Qwen3.8 Max (0902)Alibaba72
- Tencent Hy4Tencent72
- GLM-5.3Z AI71
- Inkling 2 SmallThinking Machines71
- MiniMax-M3.1-FlashMiniMax71
- Ling 3.1 Flash 1019InclusionAI69
- Grok 4.6xAI68
- Muse Spark OpenMeta68
- Qwen3.8-Flash-NextAlibaba68
- Agnes 2.5 Pro ThetaSapiens AI67
- MiMo-V2.6-ProXiaomi67
- Gemini 3.8 FlashGoogle65
- GLM-5.3-FlashZ AI64
- Mistral Large 4Mistral API and open weights64
- MiMo-V2.6-FlashXiaomi62
- Mistral Large 4 PreviewRetired public preview62
- GPT-5.6 TerraOpenAI61
- Beam 1.0Reflection58
- DeepSeek V4 Pro 0813DeepSeek58
- GPT-6 LunaOpenAI58
- K2 Horizon 375B A23BInstitute of Foundation Models58
- GPT-5.6 LunaOpenAI55
- Nanbeige4.3-33BNanbeige54
- Qwen3.8 27BAlibaba54
- JT-4.1 Flash 236B A21BChina Mobile52
- MiniMax-M3MiniMax52
- Nemotron 3.5 Super VLNVIDIA51
- Apodex 1.1Apodex46
- Solar Pro 4Upstage46
- Nemotron 3 Ultra 550B A55BNVIDIA45
- Inkling SmallThinking Machines44
- Kimi K3Kimi44
- Gemini 3.5 Flash-LiteGoogle43
- InklingThinking Machines41
- K-EXAONE 2.0LG AI Research34
- Nemotron 3.5 LightningNVIDIA33
- Command A+Cohere32
- Gemini 3.1 Flash-LiteGoogle30
- Granite 4.2 3BIBM27
- Muse GlimmerMeta27
- North Mini CodeCohere27
- Mistral Medium 3.5Mistral26
- Mistral Small 4Mistral26
- Granite 4.2 8BIBM22
- NVIDIA Nemotron 3 Nano 4BNVIDIA21
- Gemma 4 31BGoogle20
100Klu Index94
$16 / 1M · 101 tok/sPrice and speed$8 / 1M · 114 tok/s
Frontier benchmark scores
| Model | AutomationBench | CritPt v2 Hard | Omniscience v1.1 | FrontierCode 2.0 | HLE-Diamond | T-B 4.0 | T-B Science | HealthBench Professional |
|---|---|---|---|---|---|---|---|---|
| Claude Fable 5.2 | 76.4 | 76.5 | 66.77 | 49.1 | 80.2 | 61.6 | 73.3 | 68.1 |
| Claude Opus 5.5 | 69.5 | 68.0 | 48.13 | 46.4 | 67.9 | 59.6 | 61.9 | 59.9 |
| Gemini 4 Carbon | 77.8 | 69.0 | 53.5 | 40.6 | 69.1 | 65.2 | 64.8 | Not reported |
| Gemini 4 Hydrogen | 76.1 | 66.0 | 46.27 | 39.0 | 58.8 | 54.0 | 49.5 | Not reported |
| Claude Sonnet 5.5 | 71.8 | 59.0 | 30.47 | 41.7 | 53.3 | 63.6 | 53.3 | 63.0 |
| Muse Wonder Full | 71.3 | 56.5 | 30.6 | 37.8 | 48.5 | 57.1 | 37.1 | Not reported |
| Claude Fable 5.1 | 59.4 | 62.0 | 46.27 | 35.7 | 60.1 | 55.1 | 43.3 | 57.8 |
| Kimi K3.1 Pro | 73.3 | 49.5 | 29.87 | 25.5 | 37.3 | 67.2 | 34.3 | Not reported |
| Gemini 4 Argon | 77.5 | 53.5 | 46.1 | 25.7 | 56.7 | 57.1 | 52.9 | 55.1 |
| Muse Wonder | 68.1 | 51.0 | 25.1 | 27.9 | 39.0 | 52.0 | 25.7 | Not reported |
| GPT-6 Astra | 68.5 | 72.0 | 44.57 | 42.4 | 63.8 | 59.6 | 63.3 | 66.0 |
| GPT-6.1 Sol | 66.6 | 69.5 | 42.37 | 40.6 | 58.8 | 56.1 | 58.1 | 65.1 |
How the Klu Index works
The Klu Index combines accuracy, use-case evaluations, human preference, and speed into one score out of 100. The fastest way to compare models.
AccuracyArtificial Analysis Intelligence Index v5.0.1 across nine evaluations35%
EvaluationsLegal, finance, healthcare, engineering, economics, and strategy30%
PreferenceBlind pairwise judgments of finished work20%
PerformanceOutput speed and time to first token15%
Models by lab
Anthropic6 models
ModelKlu Index$ / 1M
Claude Fable 5.2100$16.00
Claude Opus 5.594$8.00
Claude Sonnet 5.590$4.00
Claude Fable 5.184$20.00
Claude Opus 580$10.00
Claude Haiku 5.572$0.20
Google7 models
ModelKlu Index$ / 1M
Gemini 4 Carbon93$4.00
Gemini 4 Hydrogen91$3.00
Gemini 4 Argon83$4.00
Gemini 3.8 Flash65$1.50
Gemini 3.5 Flash-Lite43$0.85
Gemini 3.1 Flash-Lite30$0.56
Gemma 4 31B20—
Meta6 models
ModelKlu Index$ / 1M
Muse Wonder FullMuse Code subscription87—
Muse WonderMeta Model API and Muse Code83$8.00
Muse Ember81$0.20
Muse Spark 1.376$2.00
Muse Spark Open68$2.00
Muse Glimmer27$0.53
OpenAI7 models
ModelKlu Index$ / 1M
GPT-6 Astra82$20.00
GPT-6.1 Sol82$4.00
GPT-6 Sol76$4.00
GPT-6.1 Terra76$2.00
GPT-5.6 Terra61$4.50
GPT-6 Luna58$0.20
GPT-5.6 Luna55$0.45
Alibaba3 models
ModelKlu Index$ / 1M
Qwen3.8 Max (0902)72$3.00
Qwen3.8-Flash-Next68$0.23
Qwen3.8 27B54$1.13
Thinking Machines3 models
ModelKlu Index$ / 1M
Inkling 2 Small71$0.48
Inkling Small44$0.53
Inkling41$1.76
Mistral4 models
ModelKlu Index$ / 1M
Mistral Large 4Mistral API and open weights64$1.05
Mistral Large 4 PreviewRetired public preview62$1.05
Mistral Medium 3.526$3.00
Mistral Small 426$0.26
NVIDIA4 models
ModelKlu Index$ / 1M
Nemotron 3.5 Super VL51$0.26
Nemotron 3 Ultra 550B A55B45$1.13
Nemotron 3.5 Lightning33$0.10
NVIDIA Nemotron 3 Nano 4B21—
Other labs27 models
ModelKlu Index$ / 1M
Kimi K3.1 Pro84$4.00
DeepSeek V4.1 Flash 101779$0.53
Step 5 Preview73$1.43
Grok 4.772$3.00
Tencent Hy472$1.25
GLM-5.371$2.15
MiniMax-M3.1-Flash71$0.26
Ling 3.1 Flash 101969$0.45
Grok 4.668$3.00
Agnes 2.5 Pro Theta67$0.15
MiMo-V2.6-Pro67$0.54
GLM-5.3-Flash64$0.24
MiMo-V2.6-Flash62$0.18
Beam 1.058$0.60
DeepSeek V4 Pro 081358$1.98
K2 Horizon 375B A23B58—
Nanbeige4.3-33B54—
JT-4.1 Flash 236B A21B52—
MiniMax-M352$0.53
Apodex 1.146$0.98
Solar Pro 446$0.53
Kimi K344$6.00
K-EXAONE 2.034—
Command A+32—
Granite 4.2 3B27—
North Mini Code27—
Granite 4.2 8B22$0.11
About the leaderboard
- What is the Klu Index?
- One score out of 100 that blends public benchmark accuracy, use-case evaluations, blind preference judgments, and speed. Scores are scaled so the top model lands at 100.
- How is speed measured?
- Tokens per second is the median output speed on each provider's API. First token is the median time, in seconds, before the first token arrives.
- How often is it updated?
- Each time Klu refreshes the snapshot. The date at the top of the page shows when the current numbers were pulled.
- Which model is best for me?
- The one that wins on your own evals. Klu runs every model on your evals and shows which one comes out ahead for each use case.
The leaderboard is a starting point
The model at the top here isn't always the one that wins on your work. Klu runs every model on your evals and reruns them each time a new one ships.