Skip to content
ModelCap

Charts

AI model price vs performance, and the frontier over time

Two views of the ModelCap Index, drawn from the same numbers as the main board. The first sets every current ranked model that lists an API price against its Index score and traces the best-value line: the highest measured score available at or below each price. The second follows the best measured score by the date models were first listed.

Snapshot as of 19 November 2026

Capability vs price

123 of 371 current ranked models list an API price. The best-value line runs through 11 measured models, from Llama 3.1 8B Instruct ($0.025 per million tokens, score 13.8) to Gemini 4 Carbon ($4.00, score 87.8).

ModelCap Index score against blended API price120 current ranked models. The best-value line joins 11 measured models, each the highest measured score at or below its price.0102030405060708090100$0.01$0.10$1$10$100ModelCap Index scoreBlended API price, US$ per million tokens (log scale)Ling 3.0 Tiny · #149 · score 47.2 (estimated) · $0.022 per 1M tokens blendedNex-N2.5-Mini · #40 · score 71.7 (estimated) · $0.044 per 1M tokens blendedGemma 3 4B · #267 · score 26.8 (measured) · $0.063 per 1M tokens blendedLlama 3.2 1B Instruct · #367 · score 4.8 (measured) · $0.071 per 1M tokens blendedGranite 4.2 8B · #269 · score 26.3 (measured) · $0.082 per 1M tokens blendedPhi 4 · #308 · score 19.2 (measured) · $0.088 per 1M tokens blendedNemotron 3.5 Lightning · #219 · score 34.9 (estimated) · $0.103 per 1M tokens blendedNemotron 3 Nano 30B A3B · #245 · score 31.2 (measured) · $0.105 per 1M tokens blendedMistral Small 3.2 24B · #258 · score 28.4 (estimated) · $0.106 per 1M tokens blendedLFM3-1.7B · #155 · score 44.8 (estimated) · $0.11 per 1M tokens blendedNex-N2.5-Pro · #32 · score 74.5 (estimated) · $0.119 per 1M tokens blendedLlama 3.2 3B Instruct · #358 · score 8.5 (measured) · $0.122 per 1M tokens blendedQwen3 32B · #175 · score 39.5 (measured) · $0.13 per 1M tokens blendedgpt-oss-safeguard-20b · #260 · score 28.2 (inherited) · $0.131 per 1M tokens blendedQwen3 30B A3B Instruct 2507 · #147 · score 47.7 (measured) · $0.15 per 1M tokens blendedLlama 4 Scout · #239 · score 31.8 (measured) · $0.15 per 1M tokens blendedNemotron 3 Super · #163 · score 41.7 (measured) · $0.156 per 1M tokens blendedQwen3.5-9B · #164 · score 41.1 (estimated) · $0.172 per 1M tokens blendedNex-N2.6 · #35 · score 72.9 (estimated) · $0.19 per 1M tokens blendedGPT-6 Luna Pro · #51 · score 67.6 (estimated) · $0.20 per 1M tokens blendedGPT-6 Luna · #58 · score 66.3 (measured) · $0.20 per 1M tokens blendedMuse Ember · #63 · score 65.2 (measured) · $0.20 per 1M tokens blendedClaude Haiku 5.5 · #93 · score 58.3 (measured) · $0.20 per 1M tokens blendedLFM2.5-24B-A2B · #139 · score 49.4 (estimated) · $0.212 per 1M tokens blendedQwen3 30B A3B · #222 · score 34.6 (measured) · $0.215 per 1M tokens blendedQwen3.8 Flash · #105 · score 56.1 (estimated) · $0.23 per 1M tokens blendedNemotron 3.5 Super VL · #133 · score 50.9 (measured) · $0.237 per 1M tokens blendedGemma 4 31B · #42 · score 70.7 (measured) · $0.25 per 1M tokens blendedMiniMax M3.1 Flash · #33 · score 74.1 (measured) · $0.262 per 1M tokens blendedDeepSeek V4.1 Flash 1017 · #92 · score 58.5 (measured) · $0.262 per 1M tokens blendedQwen3 Coder 30B A3B Instruct · #130 · score 51.2 (estimated) · $0.262 per 1M tokens blendedSolar Pro 4 · #145 · score 48.0 (measured) · $0.262 per 1M tokens blendedgpt-oss-120b · #174 · score 39.8 (measured) · $0.262 per 1M tokens blendedQwen3 VL 30B A3B Instruct · #184 · score 38.7 (estimated) · $0.262 per 1M tokens blendedGPT-4o-mini · #224 · score 34.4 (measured) · $0.262 per 1M tokens blendedMistral Small 4 · #279 · score 24.6 (estimated) · $0.262 per 1M tokens blendedCommand R (08-2024) · #310 · score 18.4 (measured) · $0.262 per 1M tokens blendedQwen3 Coder Next · #86 · score 59.5 (estimated) · $0.29 per 1M tokens blendedDeepSeek V3.2 · #89 · score 58.8 (measured) · $0.29 per 1M tokens blendedQwen3.6 35B A3B · #116 · score 52.6 (estimated) · $0.30 per 1M tokens blendedQwen3 235B A22B Instruct 2507 · #91 · score 58.5 (measured) · $0.305 per 1M tokens blendedGLM 4.5 Air · #154 · score 45.1 (measured) · $0.31 per 1M tokens blendedQwen3 Next 80B A3B Instruct · #124 · score 51.7 (measured) · $0.35 per 1M tokens blendedQwen2.5 72B Instruct · #259 · score 28.3 (measured) · $0.37 per 1M tokens blendedQwen3 VL 8B Instruct · #221 · score 34.7 (estimated) · $0.375 per 1M tokens blendedBeam 1.0 501B A23B · #67 · score 63.4 (measured) · $0.438 per 1M tokens blendedStep 3.7 Flash · #128 · score 51.6 (estimated) · $0.438 per 1M tokens blendedGLM 4.6V · #151 · score 46.1 (measured) · $0.45 per 1M tokens blendedDeepSeek V3.1 Terminus · #97 · score 56.9 (measured) · $0.453 per 1M tokens blendedGPT-5.4 Nano · #126 · score 51.6 (measured) · $0.463 per 1M tokens blendedInkling 2 Small · #98 · score 56.8 (measured) · $0.475 per 1M tokens blendedQwen3 Coder 480B A35B · #142 · score 48.9 (measured) · $0.475 per 1M tokens blendedLlama 4 Maverick · #226 · score 33.5 (measured) · $0.512 per 1M tokens blendedLongCat 2.0 · #23 · score 77.6 (estimated) · $0.525 per 1M tokens blendedMiniMax M3 · #66 · score 63.7 (measured) · $0.525 per 1M tokens blendedMotif 3 · #82 · score 59.9 (estimated) · $0.525 per 1M tokens blendedMiMo-V2.6-Pro · #20 · score 78.4 (measured) · $0.544 per 1M tokens blendedQwen3.7 Plus · #43 · score 70.6 (measured) · $0.56 per 1M tokens blendedQwen3.8 27B · #68 · score 63.1 (measured) · $0.595 per 1M tokens blendedCommand A+ · #223 · score 34.6 (estimated) · $0.60 per 1M tokens blendedQwen3 VL 235B A22B Instruct · #107 · score 55.6 (measured) · $0.632 per 1M tokens blendedMuse Glimmer 30B · #56 · score 66.8 (estimated) · $0.637 per 1M tokens blendedLlama 3.3 70B Instruct · #244 · score 31.4 (measured) · $0.71 per 1M tokens blendedQwen3.5-122B-A10B · #99 · score 56.7 (measured) · $0.715 per 1M tokens blendedQwen2.5 Coder 32B Instruct · #290 · score 23.3 (measured) · $0.745 per 1M tokens blendedMiMo-V2.5 · #65 · score 64.4 (measured) · $0.80 per 1M tokens blendedGemini 3.5 Flash Lite · #41 · score 71.1 (measured) · $0.85 per 1M tokens blendedNemotron 3 Ultra · #261 · score 28.0 (measured) · $0.925 per 1M tokens blendedR1 0528 · #95 · score 58.1 (measured) · $1.00 per 1M tokens blendedSeed 2.1 Turbo · #28 · score 76.0 (estimated) · $1.00 per 1M tokens blendedMistral Large 4 · #96 · score 58.1 (measured) · $1.05 per 1M tokens blendedKimi K2 Thinking · #71 · score 62.3 (measured) · $1.07 per 1M tokens blendedQwen3.5 397B A17B · #52 · score 67.2 (measured) · $1.09 per 1M tokens blendedDeepSeek V4 Pro 0813 · #34 · score 73.0 (measured) · $1.19 per 1M tokens blendedQwen3 235B A22B Thinking 2507 · #118 · score 52.3 (measured) · $1.21 per 1M tokens blendedHy4 preview · #45 · score 69.5 (measured) · $1.25 per 1M tokens blendedStep 5 Preview · #49 · score 68.4 (measured) · $1.43 per 1M tokens blendedGrok 4.20 Multi-Agent · #27 · score 76.2 (measured) · $1.56 per 1M tokens blendedGPT-5.4 Mini · #47 · score 68.5 (measured) · $1.69 per 1M tokens blendedInkling · #62 · score 65.6 (measured) · $1.76 per 1M tokens blendedMuse Spark Open · #21 · score 78.4 (estimated) · $2.00 per 1M tokens blendedGPT-6.1 Terra · #37 · score 72.5 (measured) · $2.00 per 1M tokens blendedQwen4 Max · #36 · score 72.8 (measured) · $2.14 per 1M tokens blendedKimi K3.1 Pro · #11 · score 80.0 (measured) · $2.39 per 1M tokens blendedGLM 5.3 Prime · #12 · score 79.7 (estimated) · $2.58 per 1M tokens blendedGemini 4 Hydrogen · #9 · score 80.4 (measured) · $3.00 per 1M tokens blendedGrok 4.7 · #59 · score 66.1 (measured) · $3.00 per 1M tokens blendedMistral Medium 3.5 · #87 · score 59.4 (measured) · $3.00 per 1M tokens blendedMixtral 8x22B Instruct · #318 · score 15.9 (measured) · $3.00 per 1M tokens blendedo3 · #75 · score 61.0 (measured) · $3.50 per 1M tokens blendedGemini 4 Argon · #3 · score 86.6 (measured) · $4.00 per 1M tokens blendedGPT-6.1 Sol Pro · #10 · score 80.4 (estimated) · $4.00 per 1M tokens blendedClaude Sonnet 5.5 · #24 · score 77.3 (measured) · $4.00 per 1M tokens blendedGPT-6.2 Sol · #29 · score 75.8 (measured) · $4.00 per 1M tokens blendedKimi K3 · #17 · score 79.0 (measured) · $4.14 per 1M tokens blendedCommand A · #167 · score 39.9 (measured) · $4.38 per 1M tokens blendedGPT-4o (2024-08-06) · #198 · score 36.9 (measured) · $4.38 per 1M tokens blendedCommand R+ (08-2024) · #294 · score 22.9 (measured) · $4.38 per 1M tokens blendedGemini 3.1 Pro Preview · #16 · score 79.2 (measured) · $4.50 per 1M tokens blendedGemini 4 Neon · #2 · score 87.2 (measured) · $6.00 per 1M tokens blendedQwen3.8 Max Prime · #13 · score 79.5 (estimated) · $6.00 per 1M tokens blendedClaude Opus 5.5 · #4 · score 85.8 (measured) · $8.00 per 1M tokens blendedMuse Wonder · #5 · score 85.6 (measured) · $8.00 per 1M tokens blendedGPT-5.5 · #18 · score 78.8 (measured) · $11.25 per 1M tokens blendedClaude Fable 5.2 · #6 · score 84.2 (measured) · $16.00 per 1M tokens blendedGPT-6.2 Astra · #22 · score 77.9 (measured) · $16.00 per 1M tokens blendedGPT-6 Astra Pro · #14 · score 79.4 (estimated) · $20.00 per 1M tokens blendedo3 Pro · #70 · score 62.5 (estimated) · $35.00 per 1M tokens blendedGPT-5.5 Pro · #15 · score 79.3 (measured) · $67.50 per 1M tokens blendedLlama 3.1 8B Instruct · #325 · score 13.8 (measured) · $0.025 per 1M tokens blendedgpt-oss-20b · #238 · score 32.2 (measured) · $0.036 per 1M tokens blendedGemma 3 12B · #201 · score 36.7 (measured) · $0.075 per 1M tokens blendedGemma 3 27B · #160 · score 42.7 (measured) · $0.10 per 1M tokens blendedGemma 4 26B A4B · #60 · score 65.9 (measured) · $0.121 per 1M tokens blendedMiMo-V2.6-Flash · #46 · score 68.8 (measured) · $0.175 per 1M tokens blendedGLM 5.3 Flash · #26 · score 76.4 (measured) · $0.237 per 1M tokens blendedGLM 5.3 · #19 · score 78.5 (measured) · $0.402 per 1M tokens blendedGemini 3.8 Flash · #8 · score 81.4 (measured) · $1.65 per 1M tokens blendedMuse Spark 1.3 · #7 · score 82.4 (measured) · $2.00 per 1M tokens blendedGemini 4 Carbon · #1 · score 87.8 (measured) · $4.00 per 1M tokens blendedGemini 4 CarbonMuse Spark 1.3GLM 5.3GLM 5.3 FlashMiMo-V2.6-FlashGemma 4 26B A4BGemma 3 27BGemma 3 12Bgpt-oss-20bLlama 3.1 8B Instruct
Measured score Inherited or estimated score Best measured score at or below each price

Prices are the listed API prices from OpenRouter, blended three parts input to one part output, in US dollars per million tokens. The price axis is logarithmic: each gridline is ten times the one before. Inherited and estimated scores are modeled rather than measured, so they are drawn but never form the line. Models with no listed API price, including open-weight models you can only run yourself, are not drawn. Listed free of charge, and so off the logarithmic axis: Dots3-Note Preview (67.8), Flint (61.3) and North Mini Code (32.3).

The best-value line, cheapest first

Measured models on the best-value line, cheapest first
ModelIndex scorePositionBlended / 1MInput · output / 1M
Llama 3.1 8B InstructMeta13.8#325$0.025$0.02 · $0.04
gpt-oss-20bOpenAI32.2#238$0.036$0.018 · $0.09
Gemma 3 12BGoogle36.7#201$0.075$0.05 · $0.15
Gemma 3 27BGoogle42.7#160$0.10$0.08 · $0.16
Gemma 4 26B A4BGoogle65.9#60$0.121$0.076 · $0.255
MiMo-V2.6-FlashXiaomi68.8#46$0.175$0.14 · $0.28
GLM 5.3 FlashZ.ai76.4#26$0.237$0.15 · $0.50
GLM 5.3Z.ai78.5#19$0.402$0.07 · $1.40
Gemini 3.8 FlashGoogle81.4#8$1.65$0.825 · $4.13
Muse Spark 1.3Meta82.4#7$2.00$1.25 · $4.25
Gemini 4 CarbonGoogle87.8#1$4.00$2.00 · $10.00

The frontier over time

16 models have each set a new best measured score when listed, from GPT-4 (25.4, listed 28 May 2023) to Gemini 4 Carbon (87.8, listed 10 Nov 2026).

Best measured ModelCap Index score by listing date234 measured models placed at the date each was first listed. The line steps up at each of the 16 models that set a new best score.0102030405060708090100202420252026ModelCap Index scoreDate first listedGPT-3.5 Turbo · listed 28 May 2023 · score 15.9 · all versions #458Qwen 14B Chat · listed 24 Sept 2023 · score 7.3 · all versions #514Mistral 7B Instruct v0.2 · listed 11 Dec 2023 · score 7.2 · #363SOLAR 10.7B Instruct v1.0 · listed 12 Dec 2023 · score 7.6 · #359Qwen1.5 4B Chat · listed 30 Jan 2024 · score 3.7 · #369Qwen1.5 7B Chat · listed 30 Jan 2024 · score 7.3 · #360Qwen1.5 14B Chat · listed 30 Jan 2024 · score 11.7 · #353Qwen1.5 72B Chat · listed 30 Jan 2024 · score 16.1 · #317OLMo 7B Instruct · listed 23 Feb 2024 · score 3.6 · all versions #529Mistral Large · listed 26 Feb 2024 · score 29.0 · all versions #386Claude 3 Haiku · listed 13 Mar 2024 · score 20.3 · all versions #445gemma 1.1 7b it · listed 26 Mar 2024 · score 10.5 · #354gemma 1.1 2b it · listed 26 Mar 2024 · score 5.4 · all versions #525Qwen1.5 32B Chat · listed 3 Apr 2024 · score 13.0 · #333Mixtral 8x22B Instruct · listed 17 Apr 2024 · score 15.9 · #318Phi 3 mini 4k instruct · listed 22 Apr 2024 · score 6.4 · all versions #521Phi 3 mini 128k instruct · listed 22 Apr 2024 · score 6.1 · all versions #522Phi 3 medium 4k instruct · listed 7 May 2024 · score 12.2 · #339Phi 3 small 8k instruct · listed 7 May 2024 · score 9.4 · #357GPT-4o · listed 13 May 2024 · score 34.6 · all versions #343Qwen2 72B Instruct · listed 28 May 2024 · score 19.8 · all versions #446gemma 2 9b it · listed 24 Jun 2024 · score 20.3 · #306Gemma 2 27B · listed 13 Jul 2024 · score 25.1 · all versions #408gemma 2 2b it · listed 16 Jul 2024 · score 11.7 · #352GPT-4o-mini · listed 18 Jul 2024 · score 34.4 · #224GPT-4o-mini (2024-07-18) · listed 18 Jul 2024 · score 31.6 · all versions #368Llama 3.1 70B Instruct · listed 23 Jul 2024 · score 26.0 · all versions #402Llama 3.1 8B Instruct · listed 23 Jul 2024 · score 13.8 · #325GPT-4o (2024-08-06) · listed 6 Aug 2024 · score 36.9 · #198Command R+ (08-2024) · listed 30 Aug 2024 · score 22.9 · #294Command R (08-2024) · listed 30 Aug 2024 · score 18.4 · #310Qwen2.5 72B Instruct · listed 19 Sept 2024 · score 28.3 · #259Llama 3 1 Nemotron 51B Instruct · listed 22 Sept 2024 · score 24.6 · #278Llama 3.2 3B Instruct · listed 25 Sept 2024 · score 8.5 · #358Llama 3.2 1B Instruct · listed 25 Sept 2024 · score 4.8 · #367granite 3.0 2b instruct · listed 2 Oct 2024 · score 8.2 · all versions #510granite 3.0 8b instruct · listed 2 Oct 2024 · score 11.0 · all versions #501Qwen2.5 Coder 32B Instruct · listed 11 Nov 2024 · score 23.3 · #290Llama 3.1 Tulu 3 8B · listed 20 Nov 2024 · score 14.9 · #320Llama 3.1 Tulu 3 70B · listed 20 Nov 2024 · score 24.4 · #280QwQ 32B Preview · listed 27 Nov 2024 · score 7.8 · all versions #511Llama 3.3 70B Instruct · listed 6 Dec 2024 · score 31.4 · #244granite 3.1 8b instruct · listed 6 Dec 2024 · score 13.9 · all versions #468granite 3.1 2b instruct · listed 6 Dec 2024 · score 10.9 · all versions #502DeepSeek V3 · listed 26 Dec 2024 · score 41.5 · all versions #269Phi 4 · listed 10 Jan 2025 · score 19.2 · #308R1 · listed 20 Jan 2025 · score 51.0 · all versions #216Mistral Small 3 · listed 30 Jan 2025 · score 23.6 · all versions #424o3 Mini · listed 31 Jan 2025 · score 39.2 · all versions #288Qwen-Plus · listed 1 Feb 2025 · score 38.9 · all versions #297QwQ 32B · listed 5 Mar 2025 · score 36.9 · #199Gemma 3 27B · listed 12 Mar 2025 · score 42.7 · #160Command A · listed 13 Mar 2025 · score 39.9 · #167Gemma 3 12B · listed 13 Mar 2025 · score 36.7 · #201Gemma 3 4B · listed 13 Mar 2025 · score 26.8 · #267Llama 3 3 Nemotron Super 49B v1 · listed 16 Mar 2025 · score 34.7 · all versions #340Mistral Small 3.1 24B · listed 17 Mar 2025 · score 28.6 · all versions #389DeepSeek V3 0324 · listed 24 Mar 2025 · score 50.2 · all versions #224Llama 4 Scout · listed 5 Apr 2025 · score 31.8 · #239Llama 4 Maverick · listed 5 Apr 2025 · score 33.5 · #226Llama 3 1 Nemotron Ultra 253B v1 · listed 7 Apr 2025 · score 39.1 · #180GPT-4.1 Nano · listed 14 Apr 2025 · score 32.6 · all versions #360GPT-4.1 Mini · listed 14 Apr 2025 · score 46.9 · all versions #247o4 Mini · listed 16 Apr 2025 · score 48.3 · all versions #234Qwen3 235B A22B · listed 28 Apr 2025 · score 52.4 · #117Qwen3 32B · listed 28 Apr 2025 · score 39.5 · #175Qwen3 30B A3B · listed 28 Apr 2025 · score 34.6 · #222Mistral Medium 3 · listed 7 May 2025 · score 48.2 · all versions #236Claude Sonnet 4 · listed 22 May 2025 · score 52.5 · all versions #194Claude Opus 4 · listed 22 May 2025 · score 60.6 · all versions #136R1 0528 · listed 28 May 2025 · score 58.1 · #95gemma 3n E4B it · listed 3 Jun 2025 · score 30.9 · #247Gemini 2.5 Flash · listed 17 Jun 2025 · score 52.8 · all versions #188MiniMax M1 · listed 17 Jun 2025 · score 43.8 · all versions #259Qwen3 235B A22B Instruct 2507 · listed 21 Jul 2025 · score 58.5 · #91Qwen3 Coder 480B A35B · listed 23 Jul 2025 · score 48.9 · #142Qwen3 235B A22B Thinking 2507 · listed 25 Jul 2025 · score 52.3 · #118Llama 3 3 Nemotron Super 49B v1 5 · listed 25 Jul 2025 · score 39.1 · #179GLM 4.5 Air · listed 25 Jul 2025 · score 45.1 · #154GLM 4.5 · listed 25 Jul 2025 · score 55.2 · all versions #178Qwen3 30B A3B Instruct 2507 · listed 29 Jul 2025 · score 47.7 · #147gpt-oss-20b · listed 5 Aug 2025 · score 32.2 · #238gpt-oss-120b · listed 5 Aug 2025 · score 39.8 · #174GPT-5 Nano · listed 7 Aug 2025 · score 36.6 · all versions #320GPT-5 Mini · listed 7 Aug 2025 · score 48.2 · all versions #235GPT-5 · listed 7 Aug 2025 · score 63.0 · all versions #125GLM 4.5V · listed 11 Aug 2025 · score 40.1 · all versions #273Mistral Medium 3.1 · listed 13 Aug 2025 · score 53.9 · all versions #185DeepSeek V3.1 · listed 21 Aug 2025 · score 57.0 · all versions #162Qwen3 Next 80B A3B Instruct · listed 11 Sept 2025 · score 51.7 · #124granite 4.0 h small · listed 16 Sept 2025 · score 25.2 · #273DeepSeek V3.1 Terminus · listed 22 Sept 2025 · score 56.9 · #97Qwen3 Max · listed 23 Sept 2025 · score 59.3 · all versions #148Qwen3 VL 235B A22B Instruct · listed 23 Sept 2025 · score 55.6 · #107DeepSeek V3.2 Exp · listed 29 Sept 2025 · score 60.1 · all versions #140GLM 4.6 · listed 30 Sept 2025 · score 58.5 · all versions #153GPT-5 Pro · listed 6 Oct 2025 · score 63.3 · all versions #123Claude Haiku 4.5 · listed 15 Oct 2025 · score 56.1 · all versions #173MiniMax M2 · listed 23 Oct 2025 · score 38.6 · all versions #301Kimi K2 Thinking · listed 6 Nov 2025 · score 62.3 · #71GPT-5.1 · listed 13 Nov 2025 · score 71.0 · all versions #82Olmo 3 32B Think · listed 19 Nov 2025 · score 29.6 · all versions #384DeepSeek V3.2 · listed 1 Dec 2025 · score 58.8 · #89Mistral Large 3 2512 · listed 1 Dec 2025 · score 56.6 · all versions #167GLM 4.6V · listed 8 Dec 2025 · score 46.1 · #151Olmo 3.1 32B Think · listed 10 Dec 2025 · score 24.9 · #275GPT-5.2 · listed 10 Dec 2025 · score 64.8 · all versions #117GPT-5.2 Pro · listed 10 Dec 2025 · score 65.4 · all versions #115Nemotron 3 Nano 30B A3B · listed 14 Dec 2025 · score 31.2 · #245Gemini 3 Flash Preview · listed 17 Dec 2025 · score 75.2 · all versions #57GLM 4.7 · listed 22 Dec 2025 · score 66.8 · all versions #105MiniMax M2.1 · listed 23 Dec 2025 · score 47.8 · all versions #240GLM 4.7 Flash · listed 19 Jan 2026 · score 44.0 · all versions #258Kimi K2.5 · listed 27 Jan 2026 · score 70.1 · all versions #88Step 3.5 Flash · listed 29 Jan 2026 · score 49.9 · all versions #225GLM 5 · listed 11 Feb 2026 · score 71.5 · all versions #79MiniMax M2.5 · listed 12 Feb 2026 · score 48.6 · all versions #232Qwen3.5 397B A17B · listed 16 Feb 2026 · score 67.2 · #52Claude Sonnet 4.6 · listed 17 Feb 2026 · score 76.1 · all versions #50Gemini 3.1 Pro Preview · listed 19 Feb 2026 · score 79.2 · #16Qwen3.5-Flash · listed 25 Feb 2026 · score 50.4 · all versions #221Qwen3.5-122B-A10B · listed 25 Feb 2026 · score 56.7 · #99Qwen3.5-27B · listed 25 Feb 2026 · score 53.9 · all versions #184Qwen3.5-35B-A3B · listed 25 Feb 2026 · score 50.3 · all versions #222Gemini 3.1 Flash Lite Preview · listed 3 Mar 2026 · score 62.2 · all versions #129Mercury 2 · listed 4 Mar 2026 · score 39.0 · all versions #292GPT-5.4 · listed 5 Mar 2026 · score 77.1 · all versions #43GPT-5.4 Pro · listed 5 Mar 2026 · score 77.6 · all versions #38Nemotron 3 Super · listed 11 Mar 2026 · score 41.7 · #163GPT-5.4 Mini · listed 17 Mar 2026 · score 68.5 · #47GPT-5.4 Nano · listed 17 Mar 2026 · score 51.6 · #126MiniMax M2.7 · listed 18 Mar 2026 · score 56.8 · all versions #164Grok 4.20 · listed 31 Mar 2026 · score 57.6 · all versions #160Grok 4.20 Multi-Agent · listed 31 Mar 2026 · score 76.2 · #27Trinity Large Thinking · listed 1 Apr 2026 · score 44.2 · #157Qwen3.6 Plus · listed 2 Apr 2026 · score 68.2 · all versions #97Gemma 4 31B · listed 2 Apr 2026 · score 70.7 · #42Gemma 4 26B A4B · listed 3 Apr 2026 · score 65.9 · #60granite 4.1 8b · listed 6 Apr 2026 · score 28.9 · all versions #387GLM 5.1 · listed 7 Apr 2026 · score 75.3 · all versions #56Claude Opus 4.7 · listed 16 Apr 2026 · score 83.8 · all versions #9Kimi K2.6 · listed 20 Apr 2026 · score 73.6 · all versions #64MiMo-V2.5 · listed 22 Apr 2026 · score 64.4 · #65MiMo-V2.5-Pro · listed 22 Apr 2026 · score 75.7 · all versions #54Hy3 preview · listed 22 Apr 2026 · score 54.4 · all versions #180DeepSeek V4 Flash 0423 · listed 24 Apr 2026 · score 64.5 · all versions #119DeepSeek V4 Pro 0423 · listed 24 Apr 2026 · score 72.3 · all versions #75GPT-5.5 · listed 24 Apr 2026 · score 78.8 · #18GPT-5.5 Pro · listed 24 Apr 2026 · score 79.3 · #15Qwen3.6 Max Preview · listed 27 Apr 2026 · score 73.1 · all versions #66Mistral Medium 3.5 · listed 30 Apr 2026 · score 59.4 · #87Grok 4.3 · listed 30 Apr 2026 · score 67.1 · all versions #102Gemini 3.5 Flash · listed 19 May 2026 · score 77.2 · all versions #41Qwen3.7 Max · listed 21 May 2026 · score 76.8 · all versions #46Claude Opus 4.8 · listed 27 May 2026 · score 79.1 · all versions #27MiniMax M3 · listed 31 May 2026 · score 63.7 · #66Qwen3.7 Plus · listed 3 Jun 2026 · score 70.6 · #43Nemotron 3 Ultra · listed 4 Jun 2026 · score 28.0 · #261Claude Fable 5 · listed 9 Jun 2026 · score 83.8 · all versions #8GLM 5.2 · listed 16 Jun 2026 · score 75.7 · all versions #53Claude Sonnet 5 · listed 30 Jun 2026 · score 72.7 · all versions #70Hy3 · listed 6 Jul 2026 · score 70.6 · all versions #87Grok 4.5 · listed 8 Jul 2026 · score 74.7 · all versions #59GPT-5.6 Sol · listed 9 Jul 2026 · score 79.4 · all versions #23GPT-5.6 Terra · listed 9 Jul 2026 · score 74.5 · all versions #61GPT-5.6 Luna · listed 9 Jul 2026 · score 70.9 · all versions #83Muse Spark 1.1 · listed 16 Jul 2026 · score 80.4 · all versions #15Kimi K3 · listed 16 Jul 2026 · score 79.0 · #17Inkling · listed 17 Jul 2026 · score 65.6 · #62Gemini 3.5 Flash Lite · listed 21 Jul 2026 · score 71.1 · #41Gemini 3.6 Flash · listed 21 Jul 2026 · score 78.5 · all versions #33Claude Opus 5 · listed 24 Jul 2026 · score 80.5 · all versions #14Inkling Small · listed 30 Jul 2026 · score 52.0 · all versions #201DeepSeek V4 Flash 0731 · listed 31 Jul 2026 · score 74.7 · all versions #60Qwen3.8 Max (0803) · listed 3 Aug 2026 · score 78.9 · all versions #29Muse Spark 1.2 · listed 5 Aug 2026 · score 80.9 · all versions #13granite 4.2 3b · listed 7 Aug 2026 · score 26.5 · #268granite 4.2 30b · listed 7 Aug 2026 · score 38.0 · #193Solar Pro 4 · listed 10 Aug 2026 · score 48.0 · #145Grok 4.6 · listed 12 Aug 2026 · score 70.8 · all versions #84DeepSeek V4 Pro 0813 · listed 12 Aug 2026 · score 73.0 · #34Gemini 3.7 Flash · listed 13 Aug 2026 · score 79.6 · all versions #21Dots3-Note Preview · listed 14 Aug 2026 · score 67.8 · #50Qwen3.8 27B · listed 14 Aug 2026 · score 63.1 · #68GLM 5.3 · listed 18 Aug 2026 · score 78.5 · #19GLM 5.3 Flash · listed 26 Aug 2026 · score 76.4 · #26Hy4 preview · listed 28 Aug 2026 · score 69.5 · #45Granite 4.2 8B · listed 31 Aug 2026 · score 26.3 · #269Claude Fable 5.1 · listed 1 Sept 2026 · score 83.6 · all versions #10Gemini 3.8 Flash · listed 2 Sept 2026 · score 81.4 · #8Muse Spark 1.3 · listed 2 Sept 2026 · score 82.4 · #7GPT-6 Astra · listed 4 Sept 2026 · score 78.7 · all versions #31DeepSeek V4.1 Flash · listed 10 Sept 2026 · score 77.2 · all versions #42Grok 4.7 · listed 21 Sept 2026 · score 66.1 · #59MiMo-V2.6-Pro · listed 21 Sept 2026 · score 78.4 · #20MiMo-V2.6-Flash · listed 21 Sept 2026 · score 68.8 · #46GPT-6 Sol · listed 22 Sept 2026 · score 72.6 · all versions #71GPT-6 Luna · listed 22 Sept 2026 · score 66.3 · #58Claude Sonnet 5.5 · listed 28 Sept 2026 · score 77.3 · #24GPT-6.1 Sol · listed 29 Sept 2026 · score 79.8 · all versions #19Mistral Large 4 · listed 6 Oct 2026 · score 58.1 · #96Claude Haiku 5.5 · listed 7 Oct 2026 · score 58.3 · #93Step 5 Preview · listed 8 Oct 2026 · score 68.4 · #49Muse Wonder · listed 12 Oct 2026 · score 85.6 · #5Nemotron 3.5 Super VL · listed 15 Oct 2026 · score 50.9 · #133Beam 1.0 501B A23B · listed 16 Oct 2026 · score 63.4 · #67DeepSeek V4.1 Flash 1017 · listed 17 Oct 2026 · score 58.5 · #92MiniMax M3.1 Flash · listed 19 Oct 2026 · score 74.1 · #33Kimi K3.1 Pro · listed 23 Oct 2026 · score 80.0 · #11GPT-6.1 Terra · listed 26 Oct 2026 · score 72.5 · #37Muse Ember · listed 9 Nov 2026 · score 65.2 · #63Claude Fable 5.2 · listed 9 Nov 2026 · score 84.2 · #6Inkling 2 Small · listed 9 Nov 2026 · score 56.8 · #98Gemini 4 Hydrogen · listed 12 Nov 2026 · score 80.4 · #9Qwen4 Max · listed 16 Nov 2026 · score 72.8 · #36GPT-6.2 Sol · listed 17 Nov 2026 · score 75.8 · #29GPT-6.2 Astra · listed 17 Nov 2026 · score 77.9 · #22Gemini 4 Neon · listed 18 Nov 2026 · score 87.2 · #2GPT-4 · listed 28 May 2023 · score 25.4 · all versions #405GPT-4 Turbo Preview · listed 25 Jan 2024 · score 30.8 · all versions #374GPT-4 Turbo · listed 9 Apr 2024 · score 32.8 · all versions #359GPT-4o (2024-05-13) · listed 13 May 2024 · score 38.3 · all versions #303o1 · listed 17 Dec 2024 · score 52.1 · all versions #198GPT-4.1 · listed 14 Apr 2025 · score 55.2 · all versions #177o3 · listed 16 Apr 2025 · score 61.0 · #75Gemini 2.5 Pro · listed 17 Jun 2025 · score 66.2 · all versions #110Claude Opus 4.1 · listed 5 Aug 2025 · score 71.4 · all versions #80Claude Sonnet 4.5 · listed 29 Sept 2025 · score 72.4 · all versions #74Claude Opus 4.5 · listed 24 Nov 2025 · score 76.6 · all versions #47GPT-5.2 Chat · listed 10 Dec 2025 · score 77.6 · all versions #37Claude Opus 4.6 · listed 4 Feb 2026 · score 84.0 · all versions #7Claude Opus 5.5 · listed 22 Sept 2026 · score 85.8 · #4Gemini 4 Argon · listed 13 Oct 2026 · score 86.6 · #3Gemini 4 Carbon · listed 10 Nov 2026 · score 87.8 · #1Gemini 4 CarbonClaude Opus 4.6GPT-5.2 ChatClaude Opus 4.5Claude Sonnet 4.5Claude Opus 4.1Gemini 2.5 Proo3GPT-4.1o1GPT-4o (2024-05-13)GPT-4 TurboGPT-4 Turbo PreviewGPT-4
Measured model, any version Best measured score among models listed by that date

Each dot is a measured model on the all-versions board, placed at the date it was first listed in a public catalogue, usually OpenRouter, which can trail the publisher’s announcement. Scores are today’s Index scores, so the line shows how today’s evidence ranks models by when they appeared, not what was known at the time. Superseded models stay in.

Models that set a new high

Measured models that set a new best score when listed, oldest first
First listedModelIndex scorePosition now
28 May 2023GPT-4OpenAI25.4All versions #405
25 Jan 2024GPT-4 Turbo PreviewOpenAI30.8All versions #374
9 Apr 2024GPT-4 TurboOpenAI32.8All versions #359
13 May 2024GPT-4o (2024-05-13)OpenAI38.3All versions #303
17 Dec 2024o1OpenAI52.1All versions #198
14 Apr 2025GPT-4.1OpenAI55.2All versions #177
16 Apr 2025o3OpenAI61.0#75
17 Jun 2025Gemini 2.5 ProGoogle66.2All versions #110
5 Aug 2025Claude Opus 4.1Anthropic71.4All versions #80
29 Sept 2025Claude Sonnet 4.5Anthropic72.4All versions #74
24 Nov 2025Claude Opus 4.5Anthropic76.6All versions #47
10 Dec 2025GPT-5.2 ChatOpenAI77.6All versions #37
4 Feb 2026Claude Opus 4.6Anthropic84.0All versions #7
22 Sept 2026Claude Opus 5.5Anthropic85.8#4
13 Oct 2026Gemini 4 ArgonGoogle86.6#3
10 Nov 2026Gemini 4 CarbonGoogle87.8#1

How to read these charts

Explore further