智能指数
综合全部模型按智能指数排序，共 687 个，可上下滚动
1
Claude Opus 5.5 (Max, Default Fallback)57.62Claude Sonnet 5.5 (Max, Default Fallback)56.03Claude Opus 5.5 (Xhigh, Default Fallback)56.04Claude Opus 5.5 (High, Default Fallback)53.65Claude Fable 5.1 (Max, Default Fallback)53.46Claude Fable 5.1 (Xhigh, Default Fallback)53.27
GPT-6 Astra (Max)52.78
Gemini 4 Argon (High)52.69GPT-6 Astra (Xhigh)52.410Claude Sonnet 5.5 (Xhigh, Default Fallback)51.911GPT-6.1 Sol (Max)51.812Claude Opus 5.5 (Medium, Default Fallback)51.213Claude Fable 5.1 (High, Default Fallback)51.214GPT-6.1 Sol (Xhigh)51.015GPT-6 Astra (High)50.916Claude Opus 5 (Max)50.817GPT-6.1 Sol (High)50.218Claude Opus 5 (Xhigh)49.719Claude Fable 5 (Max, Opus 4.8 Fallback)49.620GPT-6 Astra (Medium)49.621Claude Fable 5.1 (Medium, Default Fallback)48.922Claude Opus 5 (High)48.123
Muse Spark 1.3 (Max)48.124GPT-6.1 Sol (Medium)47.825GPT-6 Sol (Max)47.626GPT-5.6 Sol (Max)47.027Claude Fable 5.1 (Low, Default Fallback)46.828Claude Sonnet 5.5 (High, Default Fallback)46.829
Grok 4.7 (Xhigh)46.430Grok 4.7 (High)46.331
MiMo-V2.6-Pro46.332GPT-6 Astra (Low)45.833
Qwen3.8 Max (0902)45.434Muse Spark 1.3 (Xhigh)45.135Claude Opus 5 (Medium)44.836
GLM-5.3 (Max)44.837Grok 4.6 (High)44.338GPT-6 Sol (Xhigh)44.239Grok 4.6 (Xhigh)44.240GPT-5.6 Sol (Xhigh)44.041
Step 5 Preview43.742
Kimi K3 (Max)43.643Claude Haiku 5.5 (Max)43.444Grok 4.6 (Medium)42.845GPT-6 Sol (High)42.446GPT-5.6 Sol (High)42.347Claude Opus 5.5 (Low, Default Fallback)42.348Grok 4.7 (Low)42.249GPT-6.1 Sol (Low)42.150GPT-5.6 Terra (Max)42.151GLM-5.3-Flash41.852Claude Opus 4.8 (Max)41.853Claude Haiku 5.5 (Xhigh)41.254
Ling 3.1 Flash41.155Gemini 3.8 Flash (High)40.956Claude Sonnet 5.5 (Medium, Default Fallback)40.857Claude Opus 4.7 (Max)40.758Qwen3.8 Max40.259Qwen3.8 2.4T A95B39.960Qwen3.8-Flash-Next39.861GPT-6 Sol (Medium)39.862Gemini 3.8 Flash (Medium)39.863Gemini 3.7 Flash (Medium)39.664Muse Spark 1.2 (Xhigh)39.665
DeepSeek V4.1 Flash (Max)39.566Claude Opus 5 (Low)39.467GPT-5.6 Sol (Medium)39.268Gemini 3.7 Flash (High)39.169GPT-5.4 (Xhigh)39.070Grok 4.5 (High)38.871
Mistral Large 4 Preview38.472GPT-5.5 (Xhigh)38.473Claude Sonnet 5 (Max)38.274GPT-6 Luna (Max)38.175GPT-5.6 Terra (Xhigh)38.076MiMo-V2.6-Flash37.977Claude Haiku 5.5 (High)37.878GPT-5.6 Luna (Max)37.379GPT-5.5 (High)37.080Gemini 3.7 Flash (Low)36.981DeepSeek V4 Pro 0813 (Max)36.082Claude Sonnet 5.5 (Low, Default Fallback)35.983Grok 4.6 (Low)35.184DeepSeek V4 Flash Vision (Max)34.885GPT-6 Luna (Xhigh)34.686GPT-5.6 Luna (Xhigh)34.687Claude Haiku 5.5 (Medium)34.588Claude Sonnet 5 (Xhigh)34.489DeepSeek V4 Flash 0731 (Max)34.390GLM-5.3 (Low)34.391GPT-5.6 Terra (High)34.292GPT-6 Sol (Low)34.293Gemini 3.6 Flash (High)34.094
JT-4.1 Flash 236B A21B (Reasoning)33.995GPT-5.5 (Medium)33.896Muse Spark 1.1 (Xhigh)33.797GLM-5.2 (Max)33.798Qwen3.8 27B (Xhigh)33.799Gemini 3.5 Flash (Medium)33.6100
Motif 333.6101GPT-5.6 Sol (Low)33.5102Gemini 3.8 Flash (Low)33.5103GPT-6 Luna (High)32.9104Gemini 3.5 Flash (High)32.6105GPT-5.3 Codex (Xhigh)32.5106Motif 3 (Beta)32.3107GPT-5.6 Luna (High)32.1108Claude Opus 4.6 (Max)31.9109Claude Sonnet 5 (High)31.7110Muse Spark31.3111Claude Opus 4.7 (Non-reasoning, High)30.9112GPT-5.5 (Low)30.7113
K2 Horizon 375B A23B30.5114DeepSeek V4 Pro 0424 (Max)30.4115GPT-5.2 (Xhigh)30.4116DeepSeek V4 Pro 0424 (High)30.1117GPT-5.6 Terra (Medium)30.1118Kimi K3 (Low)30.1119Claude Sonnet 4.6 (Max)30.1120GPT-6 Luna (Medium)29.9121Gemini 3.1 Pro Preview29.7122Qwen3.7 Max29.5123Claude Haiku 5.5 (Low)29.4124
MiniMax-M329.2125Claude Opus 4.5 (Reasoning)29.1126MiMo-V2-Pro28.6127GPT-6 Sol (Non-reasoning)28.5128GPT-5.2 Codex (Xhigh)28.5129Qwen3.6 Max Preview28.4130GPT-5.6 Sol (Non-reasoning)28.3131
Nex-N2-Pro (Based on Qwen3.5-397B-A17B)28.2132
Solar Pro 428.2133Claude Sonnet 5 (Medium)28.1134Gemini 3 Pro Preview (High)28.0135GLM-5 (Reasoning)27.9136GPT-5.4 (Low)27.6137Qwen3.8 27B (Medium)27.6138GPT-5.6 Terra (Low)27.5139JT-4.1 Flash 236B A21B (Non-reasoning)27.3140Grok Build 0.1 061627.2141Qwen3.6 Plus27.0142Kimi K2.6 (Reasoning)27.0143
Quasar 438B (Max, Based on GLM-5.2)26.7144GLM-5-Turbo26.6145GPT-5.2 (Medium)26.5146
Apodex 1.126.4147Claude Opus 4.6 (Non-reasoning, High)26.4148Gemini 3 Flash Preview (Reasoning)26.3149Qwen3.8 27B (Low)26.2150GLM-5.1 (Reasoning)26.1151GPT-5.5 Instant (June 2026)26.0152MiMo-V2.5-Pro (Reasoning)26.0153Kimi K2.7 Code25.8154
Inkling Small25.7155Grok 4.20 0309 v2 (Reasoning)25.7156K2 Horizon MoVA 36B A4B25.3157
Hy325.3158Grok 4.20 0309 (Reasoning)25.2159MiMo-V2.525.2160Qwen3.7 Plus25.2161MiMo-V2-Omni-032725.1162GPT-5.6 Luna (Medium)25.0163Inkling (Xhigh)25.0164GPT-5 Codex (High)24.9165Grok 4.3 (High)24.9166Grok 4.3 (Medium)24.8167Solar Open2 250B24.7168GPT-5.1 (High)24.7169Claude Sonnet 4.6 (Non-reasoning, High)24.7170DeepSeek V4.1 Flash (Non-reasoning)24.7171Ling-3.0-flash-VL24.6172DeepSeek V4 Flash 0420 (High)24.4173Grok 4.3 (Low)24.3174Claude Sonnet 5 (Low)24.3175GLM-5.1 (Non-reasoning)24.2176DeepSeek V4 Flash 0420 (Max)24.2177GPT-5.4 mini (Xhigh)24.1178Solar Mini 424.1179MiMo-V2-Omni23.9180Gemini 3.5 Flash (Minimal)23.8181GPT-5.1 Codex (High)23.7182Claude Opus 4.5 (Non-reasoning)23.7183Kimi K2.6 (Non-reasoning)23.6184GLM 5V Turbo (Reasoning)23.5185Kimi K2.5 (Reasoning)23.5186Claude Sonnet 4.6 (Non-reasoning, Low)23.3187Claude Sonnet 5 (Non-reasoning, High)23.2188GPT-5.5 (Non-reasoning)23.2189GPT-5 (High)23.0190
Nemotron 3 Ultra 550B A55B (Reasoning)22.9191Qwen3.5 27B (Reasoning)22.9192GPT-5 (Medium)22.9193Claude 4.1 Opus (Reasoning)22.8194MiniMax-M2.522.8195MiniMax-M2.722.8196Hy3-preview (Reasoning)22.7197GPT-5.5 Instant (May 2026)22.7198Ling-3.0-flash-Fin22.6199Grok 422.5200MiMo-V2-Flash (Feb 2026)22.4201GLM-5.2 (Non-reasoning)22.4202Gemini 3 Pro Preview (Low)22.3203GLM-4.7 (Reasoning)22.2204Gemini 3.5 Flash-Lite22.2205Kimi K2 Thinking22.0206o3-pro21.9207
G9v3-39A5B21.8208GLM-5 (Non-reasoning)21.8209
KAT Coder Pro V221.7210GPT-6 Luna (Low)21.5211DeepSeek V3.2 (Reasoning)21.5212SA.X-K221.5213Qwen3.5 397B A17B (Non-reasoning)21.4214Qwen3.6 27B (Reasoning)21.4215Qwen3 Max Thinking21.3216GPT-5.6 Luna (Low)21.0217MiniMax-M2.120.9218DeepSeek V4 Pro 0424 (Non-reasoning)20.8219MiMo-V2-Flash (Reasoning)20.8220GPT-5 (Low)20.8221GPT-5.6 Terra (Non-reasoning)20.8222GPT-5.4 nano (Xhigh)20.7223Claude 4.5 Sonnet (Reasoning)20.7224Claude 4 Opus (Reasoning)20.6225GPT-5 mini (Medium)20.6226K2 Horizon 7B20.6227Qwen3.5 Omni Plus20.4228GPT-5.1 Codex mini (High)20.4229Grok 4.1 Fast (Reasoning)20.4230DeepSeek V4 Pro 0813 (Non-reasoning)20.4231o320.2232Qwen3.8 27B (Non-reasoning)20.2233Ling 3.0 Flash20.1234GPT-5.4 nano (Medium)20.0235Qwen3.6 27B (Non-reasoning)19.8236GPT-5.4 mini (Medium)19.7237
K-EXAONE 2.0 080319.7238Step 3.7 Flash19.5239Kimi K2.5 (Non-reasoning)19.4240Qwen3.5 27B (Non-reasoning)19.4241Claude 4.5 Sonnet (Non-reasoning)19.3242Qwen3.5 35B A3B (Reasoning)19.3243
LongCat 2.019.1244Claude 4 Sonnet (Reasoning)18.9245DeepSeek V4 Flash 0420 (Non-reasoning)18.9246JT-35B-Flash18.7247MiniMax-M218.6248KAT-Coder-Pro V118.6249Claude 4.1 Opus (Non-reasoning)18.6250GLM-4.6 (Reasoning)18.5251GPT-6 Luna (Non-reasoning)18.5252Qwen3.5 397B A17B (Reasoning)18.4253MiMo-V2.5-Pro (Non-reasoning)18.3254Qwen3.6 35B A3B (Reasoning)18.2255GPT-5.4 (Non-reasoning)18.2256Grok 4 Fast (Reasoning)17.9257Gemini 3 Flash Preview (Non-reasoning)17.9258Qwen3.5 122B A10B (Non-reasoning)17.7259Claude 3.7 Sonnet (Reasoning)17.7260Muse Glimmer (High)17.5261GLM-4.7 (Non-reasoning)17.4262Hy3-preview (Non-reasoning)17.0263Ling-2.6-1T17.0264GPT-5.2 (Non-reasoning)17.0265Step 3.5 Flash 260317.0266
Doubao Seed Code16.9267Claude 4.5 Haiku (Reasoning)16.9268GPT-5 mini (High)16.8269Gemma 4 26B A4B (Reasoning)16.7270o4-mini (High)16.7271Ring-2.6-1T16.6272Step 3.5 Flash16.6273Claude 4 Opus (Non-reasoning)16.6274Claude 4 Sonnet (Non-reasoning)16.6275DeepSeek V3.2 Exp (Reasoning)16.6276Qwen3 Max Thinking (Preview)16.3277Gemini 2.5 Pro16.1278MiMo-V2-Flash (Non-reasoning)16.0279DeepSeek V3.2 (Non-reasoning)16.0280K2 Horizon 3.7B15.6281Qwen3 Max15.6282Qwen3.5 122B A10B (Reasoning)15.6283Gemini 3.1 Flash-Lite15.6284GPT-5.6 Luna (Non-reasoning)15.5285Gemini 2.5 Flash Preview (Sep '25) (Reasoning)15.5286Claude 4.5 Haiku (Non-reasoning)15.4287Kimi K2 090515.3288Claude 3.7 Sonnet (Non-reasoning)15.3289Qwen3.6 35B A3B (Non-reasoning)15.2290o115.2291Qwen3.5 35B A3B (Non-reasoning)15.1292Gemini 2.5 Pro Preview (Mar' 25)15.0293GLM-4.6 (Non-reasoning)14.9294GLM-4.7-Flash (Reasoning)14.9295
Granite 4.2 30B14.8296DeepSeek V3.1 Terminus (Reasoning)14.8297Gemma 4 31B (Reasoning)14.7298Grok 3 mini Reasoning (High)14.6299Grok 4.20 0309 (Non-reasoning)14.6300Gemini 2.5 Pro Preview (May' 25)14.5301DeepSeek V3.2 Speciale14.5302K-EXAONE (Reasoning)14.4303
ERNIE 5.0 Thinking Preview14.3304Grok 4.20 0309 v2 (Non-reasoning)14.2305Mistral Medium 3.514.2306Gemma 4 12B (Reasoning)14.2307
Nova 2.0 Pro Preview (Medium)14.2308Grok Code Fast 114.1309Grok 4.3 (Non-reasoning)14.0310DeepSeek V3.1 Terminus (Non-reasoning)13.9311Gemma 4 31B (Non-reasoning)13.9312DeepSeek V3.2 Exp (Non-reasoning)13.9313
Apriel-v1.5-15B-Thinker13.8314
Mercury 213.8315DeepSeek V3.1 (Non-reasoning)13.7316Nova 2.0 Omni (Medium)13.6317DeepSeek V3.1 (Reasoning)13.5318Qwen3 VL 235B A22B (Reasoning)13.4319Apriel-v1.6-15B-Thinker13.4320Nova 2.0 Lite (High)13.4321GPT-5.1 (Non-reasoning)13.3322Qwen3.5 9B (Non-reasoning)13.3323EXAONE 4.5 33B (Reasoning)13.2324
Command A+13.1325Gemma 4 26B A4B (Non-reasoning)13.1326Qwen3.5 4B (Reasoning)13.1327DeepSeek R1 0528 (May '25)13.1328Gemini 2.5 Flash (Reasoning)13.1329GPT-5 nano (High)13.0330Nemotron 3.5 Lightning12.9331Nemotron 3 Super 120B A12B (Reasoning)12.8332Nova 2.0 Pro Preview (Low)12.8333GLM-4.5 (Reasoning)12.8334Kimi K212.7335Qwen3 235B A22B 2507 (Reasoning)12.7336GPT-4.112.7337Qwen3 Max (Preview)12.6338Nova 2.0 Lite (Medium)12.5339GPT-5 nano (Medium)12.5340Qwen3.5 Omni Flash12.5341o3-mini12.5342
MiniCPM5-2B12.5343o1-pro12.4344Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)12.4345Mercury 2.512.3346JT-MINI12.2347Grok 312.1348Seed-OSS-36B-Instruct12.1349Qwen3 235B A22B 2507 Instruct12.0350Qwen3 Coder 480B A35B Instruct11.9351Qwen3 VL 32B (Reasoning)11.9352Magistral Medium 1.211.8353
Sonar Reasoning Pro11.8354Nova 2.0 Lite (Low)11.8355HyperNova 60B 2605 (High, Based on gpt-oss-120b)11.7356MiniMax M1 80k11.7357GPT-5.4 nano (Non-reasoning)11.7358Nemotron Cascade 2 30B A3B11.7359Gemini 2.5 Flash Preview (Reasoning)11.7360gpt-oss-120b (High)11.6361K2 Think V211.5362LongCat Flash Lite11.5363GPT-5 (Minimal)11.4364DeepSeek R1 (Jan '25)11.4365o1-preview11.4366
HyperCLOVA X SEED Think (32B)11.4367Grok 4.1 Fast (Non-reasoning)11.3368Mistral Small 4 (Reasoning)11.3369GLM-4.6V (Reasoning)11.2370K-EXAONE (Non-reasoning)11.2371Qwen3 Next 80B A3B (Reasoning)11.2372Qwen3.5 9B (Reasoning)11.2373GPT-5.4 mini (Non-reasoning)11.1374Nova 2.0 Omni (Low)11.1375GLM-4.5-Air11.1376Granite 4.2 8B11.1377Grok 4 Fast (Non-reasoning)11.1378Ling 3.0 Tiny11.1379
Mi:dm K 2.5 Pro11.0380o3-mini (High)11.0381Ring-1T10.9382G9v3-3B10.8383
Trinity Large Thinking10.8384Qwen3.5 4B (Non-reasoning)10.8385
INTELLECT-3 (Based on GLM-4.5-Air)10.6386GLM-4.7-Flash (Non-reasoning)10.6387GPT-5 (ChatGPT)10.4388Solar Open 100B (Reasoning)10.4389Grok 3 Reasoning Beta10.4390Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)10.4391Nemotron 3 Nano Omni 30B A3B Reasoning10.3392gpt-oss-120b (Low)10.2393GPT-4.1 mini10.2394Llama 4 Maverick10.0395MiniMax M1 40k10.0396Nova 2.0 Pro Preview (Non-reasoning)10.0397gpt-oss-20b (Low)10.0398Qwen3 VL 235B A22B Instruct9.9399North Mini Code9.9400GPT-5 mini (Minimal)9.9401K2-V2 (High)9.9402Gemini 2.5 Flash (Non-reasoning)9.9403Qwen3 30B A3B 2507 (Reasoning)9.8404o1-mini9.8405DeepSeek V3 03249.7406Ling 2.6 Flash9.7407Qwen3 Next 80B A3B Instruct9.6408
Tri-21B-think Preview9.6409Qwen3 Coder 30B A3B Instruct9.6410GPT-4.5 (Preview)9.6411DiffusionGemma 26B A4B9.5412Qwen3 235B A22B (Reasoning)9.5413QwQ 32B9.5414Qwen3 VL 30B A3B (Reasoning)9.5415Gemini 2.0 Flash Thinking Experimental (Jan '25)9.4416Gemma 4 12B (Non-reasoning)9.4417Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)9.3418Mistral Large 39.3419Qwen3 Coder Next9.2420Motif-2-12.7B-Reasoning9.2421Mistral Medium 3.19.2422Ling-1T9.2423Nova Premier9.2424Solar Pro 2 (Preview) (Reasoning)9.1425Granite 4.2 3B9.1426Magistral Medium 19.1427Mistral Medium 39.0428K2-V2 (Medium)9.0429Llama Nemotron Super 49B v1.5 (Reasoning)9.0430Devstral Medium9.0431Mistral Small 4 (Non-reasoning)9.0432Tri-21B-Think9.0433gpt-oss-20b (High)9.0434GPT-4o (March 2025, chatgpt-4o-latest)9.0435Gemini 2.0 Flash (Feb '25)8.9436Claude 3.5 Haiku8.9437Llama 3.3 Nemotron Super 49B v1 (Reasoning)8.9438Gemma 4 E4B (Reasoning)8.9439NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)8.9440Qwen3 4B 2507 (Reasoning)8.8441MiniCPM5-1B (Reasoning)8.8442
Sarvam 105B (High)8.8443Gemini 2.0 Pro Experimental (Feb '25)8.7444Nova 2.0 Lite (Non-reasoning)8.7445Devstral Small (May '25)8.7446Claude 3 Opus8.7447MiniCPM5-1B (Non-reasoning)8.7448Sonar Reasoning8.7449Gemini 2.5 Flash Preview (Non-reasoning)8.7450Devstral 28.6451Magistral Small 1.28.6452Qwen3 32B (Reasoning)8.6453Gemini 2.5 Flash-Lite (Reasoning)8.5454DeepSeek V3 (Dec '24)8.5455GPT-4o (Nov '24)8.4456
Nanbeige4.1-3B8.4457
LFM2.5-2.6B8.4458Qwen3 VL 32B Instruct8.4459DeepSeek R1 Distill Qwen 32B8.4460GLM-4.6V (Non-reasoning)8.4461Qwen3 235B A22B (Non-reasoning)8.3462Mistral Small 3.28.2463Magistral Small 18.2464Gemini 2.0 Flash (experimental)8.2465EXAONE 4.0 32B (Reasoning)8.2466Qwen3 VL 8B (Reasoning)8.2467Qwen3 14B (Reasoning)8.2468Nova 2.0 Omni (Non-reasoning)8.2469DeepSeek R1 0528 Qwen3 8B8.1470Llama 4 Scout8.1471Qwen2.5 Max8.0472Qwen3 VL 30B A3B Instruct7.9473
Hermes 4 - Llama-3.1 70B (Reasoning)7.9474Gemini 1.5 Pro (Sep '24)7.9475Solar Pro 2 (Preview) (Non-reasoning)7.9476DeepSeek R1 Distill Llama 70B7.9477Claude 3.5 Sonnet (Oct '24)7.9478DeepSeek R1 Distill Qwen 14B7.8479
Falcon-H1R-7B7.8480GPT-4.1 nano7.8481Solar Pro 37.8482Ling-flash-2.07.8483Gemma 4 E2B (Reasoning)7.8484Qwen3 Omni 30B A3B (Reasoning)7.8485GPT-4o (Aug '24)7.7486Qwen2.5 Instruct 72B7.7487Sonar7.7488Step3 VL 10B7.7489Llama 3.3 Instruct 70B7.7490Qwen3 30B A3B (Reasoning)7.6491Sonar Pro7.6492Devstral Small (Jul '25)7.6493QwQ 32B-Preview7.6494GLM-4.5V (Reasoning)7.6495Mistral Large 2 (Nov '24)7.6496Devstral Small 27.5497Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)7.5498Qwen3 30B A3B 2507 Instruct7.5499ERNIE 4.5 300B A47B7.5500Hermes 4 - Llama-3.1 405B (Reasoning)7.5501Solar Pro 2 (Reasoning)7.5502NVIDIA Nemotron Nano 12B v2 VL (Reasoning)7.5503Gemma 4 E4B (Non-reasoning)7.5504Granite 4.1 30B7.4505NVIDIA Nemotron Nano 9B V2 (Reasoning)7.4506Hermes 4 - Llama-3.1 405B (Non-reasoning)7.4507Gemini 2.0 Flash-Lite (Feb '25)7.4508Llama Nemotron Super 49B v1.5 (Non-reasoning)7.4509Qwen3 32B (Non-reasoning)7.3510GPT-4o (May '24)7.3511Gemini 2.0 Flash-Lite (Preview)7.3512K2-V2 (Low)7.3513Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)7.3514Kimi Linear 48B A3B Instruct7.3515Llama 3.1 Instruct 405B7.3516Qwen3 8B (Reasoning)7.3517Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)7.3518Qwen3 VL 8B Instruct7.3519Qwen3 4B (Reasoning)7.2520LFM2.5-8B-A1B7.2521Claude 3.5 Sonnet (June '24)7.2522
Llama 3.1 Tulu3 405B7.2523GPT-4o (ChatGPT)7.2524Ring-flash-2.07.2525Pixtral Large7.1526Olmo 3.1 32B Think7.1527Mistral Small 3.17.1528Grok 2 (Dec '24)7.1529GPT-5 nano (Minimal)7.1530Gemini 1.5 Flash (Sep '24)7.1531Qwen3 VL 4B (Reasoning)7.0532GPT-4 Turbo7.0533Solar Pro 2 (Non-reasoning)7.0534Nova Pro7.0535Command A7.0536Qwen3.5 2B (Reasoning)6.9537Llama 3.1 Nemotron Instruct 70B6.9538Llama 3.1 Instruct 8B6.9539Grok Beta6.9540Qwen2.5 Instruct 32B6.9541NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)6.8542NVIDIA Nemotron Nano 9B V2 (Non-reasoning)6.8543Mistral Large 2 (Jul '24)6.8544Qwen3 4B 2507 Instruct6.7545Qwen2.5 Coder Instruct 32B6.7546Qwen3 14B (Non-reasoning)6.7547GPT-46.7548GLM-4.5V (Non-reasoning)6.7549Mistral Small 36.7550Gemini 2.5 Flash-Lite (Non-reasoning)6.7551Nova Lite6.7552GPT-4o mini6.7553Hermes 4 - Llama-3.1 70B (Non-reasoning)6.7554Qwen3 30B A3B (Non-reasoning)6.6555DeepSeek-V2.5 (Dec '24)6.6556Qwen3 4B (Non-reasoning)6.6557Llama 3.1 Instruct 70B6.6558Granite 4.1 8B6.6559Sarvam 30B (High)6.6560Gemini 2.0 Flash Thinking Experimental (Dec '24)6.6561DeepSeek-V2.56.6562Olmo 3.1 32B Instruct6.5563Mistral Saba6.5564DeepSeek R1 Distill Llama 8B6.5565Gemma 4 E2B (Non-reasoning)6.5566Olmo 3 32B Think6.5567Gemini 1.5 Pro (May '24)6.4568R1 17766.4569Qwen2.5 Turbo6.4570
Reka Flash (Sep '24)6.4571Llama 3.2 Instruct 90B (Vision)6.4572Solar Mini6.4573
Celeris-16.3574Grok-16.3575Qwen2 Instruct 72B6.3576
Phi-4 Mini Instruct6.3577EXAONE 4.0 32B (Non-reasoning)6.3578Qwen3.5 2B (Non-reasoning)6.2579Gemini 1.5 Flash-8B6.2580Qwen3.5 0.8B (Reasoning)6.1581DeepHermes 3 - Mistral 24B Preview (Non-reasoning)6.1582
Jamba 1.7 Large6.1583Granite 4.0 H Small6.0584Ministral 3 14B6.0585Jamba 1.5 Large6.0586Qwen3 Omni 30B A3B Instruct6.0587Hermes 3 - Llama-3.1 70B6.0588Qwen3 8B (Non-reasoning)6.0589DeepSeek-Coder-V26.0590OLMo 2 32B6.0591Jamba 1.6 Large6.0592LFM2 24B A2B5.9593Gemini 1.5 Flash (May '24)5.9594Phi-45.9595Claude 3 Sonnet5.9596Nova Micro5.9597Granite 4.1 3B5.9598Mistral Small (Sep '24)5.8599Gemini 1.0 Ultra5.8600Phi-3 Mini Instruct 3.8B5.8601NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)5.8602Gemma 3n E4B Instruct Preview (May '25)5.8603Phi-4 Multimodal Instruct5.8604Qwen2.5 Coder Instruct 7B5.8605Mistral Large (Feb '24)5.8606Mixtral 8x22B Instruct5.7607Llama 2 Chat 7B5.7608Llama 3.2 Instruct 3B5.7609MiniCPM-V 4.6 1.3B5.7610Jamba Reasoning 3B5.7611Qwen3 VL 4B Instruct5.7612Qwen1.5 Chat 110B5.7613Reka Flash 35.6614Olmo 3 7B Think5.6615Claude 2.15.6616Claude 3 Haiku5.6617OLMo 2 7B5.6618Molmo 7B-D5.6619Ling-mini-2.05.5620DeepSeek R1 Distill Qwen 1.5B5.5621Claude 2.05.5622DeepSeek-V2-Chat5.5623Mistral Small (Feb '24)5.5624Mistral Medium5.5625GPT-3.5 Turbo5.5626Ministral 3 8B5.5627Llama 3 Instruct 70B5.5628
Arctic Instruct5.4629Qwen Chat 72B5.4630LFM 40B5.4631Llama 3.2 Instruct 11B (Vision)5.4632Qwen3.5 0.8B (Non-reasoning)5.4633PALM-25.4634Gemini 1.0 Pro5.3635DeepSeek Coder V2 Lite Instruct5.3636Sarvam M (Reasoning, Based on Mistral Small 3.1)5.3637DeepSeek LLM 67B Chat (V1)5.3638Llama 2 Chat 70B5.3639Llama 2 Chat 13B5.3640Command-R+ (Apr '24)5.3641
OpenChat 3.5 (1210)5.3642
DBRX Instruct5.3643Exaone 4.0 1.2B (Reasoning)5.3644Olmo 3 7B Instruct5.2645Exaone 4.0 1.2B (Non-reasoning)5.2646LFM2.5-1.2B-Thinking5.2647Jamba 1.7 Mini5.2648LFM2 2.6B5.2649LFM2.5-1.2B-Instruct5.2650Jamba 1.5 Mini5.2651Granite 4.0 H 1B5.2652Qwen3 1.7B (Reasoning)5.2653Jamba 1.6 Mini5.2654Mixtral 8x7B Instruct5.1655Gemma 3 270M5.1656
Apertus 70B Instruct5.1657Granite 4.0 Micro5.1658DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)5.1659Claude Instant5.0660Command-R (Mar '24)5.0661Llama 65B5.0662Mistral 7B Instruct5.0663Qwen Chat 14B5.0664Granite 4.0 1B5.0665Molmo2-8B5.0666LFM2 8B A1B4.9667Granite 3.3 8B (Non-reasoning)4.9668Qwen3 1.7B (Non-reasoning)4.9669Gemma 3 27B Instruct4.9670Ministral 3 3B4.8671Apertus 8B Instruct4.8672Gemma 3 1B Instruct4.8673Gemma 3 4B Instruct4.8674Gemma 3n E2B Instruct4.8675Gemma 3n E4B Instruct4.8676Granite 4.0 350M4.8677Granite 4.0 H 350M4.8678LFM2 1.2B4.8679LFM2.5-VL-1.6B4.8680Llama 3 Instruct 8B4.8681Llama 3.2 Instruct 1B4.8682Qwen3 0.6B (Non-reasoning)4.8683Qwen3 0.6B (Reasoning)4.8684Tiny Aya Global4.8685NVIDIA Nemotron 3 Nano 4B4.8686Gemma 3 12B Instruct3.8687K2 Horizon 0.9B3.0
Claude Opus 5.5 (Max, Default Fallback)57.62Claude Sonnet 5.5 (Max, Default Fallback)56.03Claude Opus 5.5 (Xhigh, Default Fallback)56.04Claude Opus 5.5 (High, Default Fallback)53.65Claude Fable 5.1 (Max, Default Fallback)53.46Claude Fable 5.1 (Xhigh, Default Fallback)53.27GPT-6 Astra (Max)52.78Gemini 4 Argon (High)52.69GPT-6 Astra (Xhigh)52.410Claude Sonnet 5.5 (Xhigh, Default Fallback)51.911GPT-6.1 Sol (Max)51.812Claude Opus 5.5 (Medium, Default Fallback)51.213Claude Fable 5.1 (High, Default Fallback)51.214GPT-6.1 Sol (Xhigh)51.015GPT-6 Astra (High)50.916Claude Opus 5 (Max)50.817GPT-6.1 Sol (High)50.218Claude Opus 5 (Xhigh)49.719Claude Fable 5 (Max, Opus 4.8 Fallback)49.620GPT-6 Astra (Medium)49.621Claude Fable 5.1 (Medium, Default Fallback)48.922Claude Opus 5 (High)48.123Muse Spark 1.3 (Max)48.124GPT-6.1 Sol (Medium)47.825GPT-6 Sol (Max)47.626GPT-5.6 Sol (Max)47.027Claude Fable 5.1 (Low, Default Fallback)46.828Claude Sonnet 5.5 (High, Default Fallback)46.829Grok 4.7 (Xhigh)46.430Grok 4.7 (High)46.331MiMo-V2.6-Pro46.332GPT-6 Astra (Low)45.833Qwen3.8 Max (0902)45.434Muse Spark 1.3 (Xhigh)45.135Claude Opus 5 (Medium)44.836GLM-5.3 (Max)44.837Grok 4.6 (High)44.338GPT-6 Sol (Xhigh)44.239Grok 4.6 (Xhigh)44.240GPT-5.6 Sol (Xhigh)44.041Step 5 Preview43.742Kimi K3 (Max)43.643Claude Haiku 5.5 (Max)43.444Grok 4.6 (Medium)42.845GPT-6 Sol (High)42.446GPT-5.6 Sol (High)42.347Claude Opus 5.5 (Low, Default Fallback)42.348Grok 4.7 (Low)42.249GPT-6.1 Sol (Low)42.150GPT-5.6 Terra (Max)42.151GLM-5.3-Flash41.852Claude Opus 4.8 (Max)41.853Claude Haiku 5.5 (Xhigh)41.254Ling 3.1 Flash41.155Gemini 3.8 Flash (High)40.956Claude Sonnet 5.5 (Medium, Default Fallback)40.857Claude Opus 4.7 (Max)40.758Qwen3.8 Max40.259Qwen3.8 2.4T A95B39.960Qwen3.8-Flash-Next39.861GPT-6 Sol (Medium)39.862Gemini 3.8 Flash (Medium)39.863Gemini 3.7 Flash (Medium)39.664Muse Spark 1.2 (Xhigh)39.665DeepSeek V4.1 Flash (Max)39.566Claude Opus 5 (Low)39.467GPT-5.6 Sol (Medium)39.268Gemini 3.7 Flash (High)39.169GPT-5.4 (Xhigh)39.070Grok 4.5 (High)38.871Mistral Large 4 Preview38.472GPT-5.5 (Xhigh)38.473Claude Sonnet 5 (Max)38.274GPT-6 Luna (Max)38.175GPT-5.6 Terra (Xhigh)38.076MiMo-V2.6-Flash37.977Claude Haiku 5.5 (High)37.878GPT-5.6 Luna (Max)37.379GPT-5.5 (High)37.080Gemini 3.7 Flash (Low)36.981DeepSeek V4 Pro 0813 (Max)36.082Claude Sonnet 5.5 (Low, Default Fallback)35.983Grok 4.6 (Low)35.184DeepSeek V4 Flash Vision (Max)34.885GPT-6 Luna (Xhigh)34.686GPT-5.6 Luna (Xhigh)34.687Claude Haiku 5.5 (Medium)34.588Claude Sonnet 5 (Xhigh)34.489DeepSeek V4 Flash 0731 (Max)34.390GLM-5.3 (Low)34.391GPT-5.6 Terra (High)34.292GPT-6 Sol (Low)34.293Gemini 3.6 Flash (High)34.094JT-4.1 Flash 236B A21B (Reasoning)33.995GPT-5.5 (Medium)33.896Muse Spark 1.1 (Xhigh)33.797GLM-5.2 (Max)33.798Qwen3.8 27B (Xhigh)33.799Gemini 3.5 Flash (Medium)33.6100Motif 333.6101GPT-5.6 Sol (Low)33.5102Gemini 3.8 Flash (Low)33.5103GPT-6 Luna (High)32.9104Gemini 3.5 Flash (High)32.6105GPT-5.3 Codex (Xhigh)32.5106Motif 3 (Beta)32.3107GPT-5.6 Luna (High)32.1108Claude Opus 4.6 (Max)31.9109Claude Sonnet 5 (High)31.7110Muse Spark31.3111Claude Opus 4.7 (Non-reasoning, High)30.9112GPT-5.5 (Low)30.7113K2 Horizon 375B A23B30.5114DeepSeek V4 Pro 0424 (Max)30.4115GPT-5.2 (Xhigh)30.4116DeepSeek V4 Pro 0424 (High)30.1117GPT-5.6 Terra (Medium)30.1118Kimi K3 (Low)30.1119Claude Sonnet 4.6 (Max)30.1120GPT-6 Luna (Medium)29.9121Gemini 3.1 Pro Preview29.7122Qwen3.7 Max29.5123Claude Haiku 5.5 (Low)29.4124MiniMax-M329.2125Claude Opus 4.5 (Reasoning)29.1126MiMo-V2-Pro28.6127GPT-6 Sol (Non-reasoning)28.5128GPT-5.2 Codex (Xhigh)28.5129Qwen3.6 Max Preview28.4130GPT-5.6 Sol (Non-reasoning)28.3131Nex-N2-Pro (Based on Qwen3.5-397B-A17B)28.2132Solar Pro 428.2133Claude Sonnet 5 (Medium)28.1134Gemini 3 Pro Preview (High)28.0135GLM-5 (Reasoning)27.9136GPT-5.4 (Low)27.6137Qwen3.8 27B (Medium)27.6138GPT-5.6 Terra (Low)27.5139JT-4.1 Flash 236B A21B (Non-reasoning)27.3140Grok Build 0.1 061627.2141Qwen3.6 Plus27.0142Kimi K2.6 (Reasoning)27.0143Quasar 438B (Max, Based on GLM-5.2)26.7144GLM-5-Turbo26.6145GPT-5.2 (Medium)26.5146Apodex 1.126.4147Claude Opus 4.6 (Non-reasoning, High)26.4148Gemini 3 Flash Preview (Reasoning)26.3149Qwen3.8 27B (Low)26.2150GLM-5.1 (Reasoning)26.1151GPT-5.5 Instant (June 2026)26.0152MiMo-V2.5-Pro (Reasoning)26.0153Kimi K2.7 Code25.8154Inkling Small25.7155Grok 4.20 0309 v2 (Reasoning)25.7156K2 Horizon MoVA 36B A4B25.3157Hy325.3158Grok 4.20 0309 (Reasoning)25.2159MiMo-V2.525.2160Qwen3.7 Plus25.2161MiMo-V2-Omni-032725.1162GPT-5.6 Luna (Medium)25.0163Inkling (Xhigh)25.0164GPT-5 Codex (High)24.9165Grok 4.3 (High)24.9166Grok 4.3 (Medium)24.8167Solar Open2 250B24.7168GPT-5.1 (High)24.7169Claude Sonnet 4.6 (Non-reasoning, High)24.7170DeepSeek V4.1 Flash (Non-reasoning)24.7171Ling-3.0-flash-VL24.6172DeepSeek V4 Flash 0420 (High)24.4173Grok 4.3 (Low)24.3174Claude Sonnet 5 (Low)24.3175GLM-5.1 (Non-reasoning)24.2176DeepSeek V4 Flash 0420 (Max)24.2177GPT-5.4 mini (Xhigh)24.1178Solar Mini 424.1179MiMo-V2-Omni23.9180Gemini 3.5 Flash (Minimal)23.8181GPT-5.1 Codex (High)23.7182Claude Opus 4.5 (Non-reasoning)23.7183Kimi K2.6 (Non-reasoning)23.6184GLM 5V Turbo (Reasoning)23.5185Kimi K2.5 (Reasoning)23.5186Claude Sonnet 4.6 (Non-reasoning, Low)23.3187Claude Sonnet 5 (Non-reasoning, High)23.2188GPT-5.5 (Non-reasoning)23.2189GPT-5 (High)23.0190Nemotron 3 Ultra 550B A55B (Reasoning)22.9191Qwen3.5 27B (Reasoning)22.9192GPT-5 (Medium)22.9193Claude 4.1 Opus (Reasoning)22.8194MiniMax-M2.522.8195MiniMax-M2.722.8196Hy3-preview (Reasoning)22.7197GPT-5.5 Instant (May 2026)22.7198Ling-3.0-flash-Fin22.6199Grok 422.5200MiMo-V2-Flash (Feb 2026)22.4201GLM-5.2 (Non-reasoning)22.4202Gemini 3 Pro Preview (Low)22.3203GLM-4.7 (Reasoning)22.2204Gemini 3.5 Flash-Lite22.2205Kimi K2 Thinking22.0206o3-pro21.9207G9v3-39A5B21.8208GLM-5 (Non-reasoning)21.8209KAT Coder Pro V221.7210GPT-6 Luna (Low)21.5211DeepSeek V3.2 (Reasoning)21.5212SA.X-K221.5213Qwen3.5 397B A17B (Non-reasoning)21.4214Qwen3.6 27B (Reasoning)21.4215Qwen3 Max Thinking21.3216GPT-5.6 Luna (Low)21.0217MiniMax-M2.120.9218DeepSeek V4 Pro 0424 (Non-reasoning)20.8219MiMo-V2-Flash (Reasoning)20.8220GPT-5 (Low)20.8221GPT-5.6 Terra (Non-reasoning)20.8222GPT-5.4 nano (Xhigh)20.7223Claude 4.5 Sonnet (Reasoning)20.7224Claude 4 Opus (Reasoning)20.6225GPT-5 mini (Medium)20.6226K2 Horizon 7B20.6227Qwen3.5 Omni Plus20.4228GPT-5.1 Codex mini (High)20.4229Grok 4.1 Fast (Reasoning)20.4230DeepSeek V4 Pro 0813 (Non-reasoning)20.4231o320.2232Qwen3.8 27B (Non-reasoning)20.2233Ling 3.0 Flash20.1234GPT-5.4 nano (Medium)20.0235Qwen3.6 27B (Non-reasoning)19.8236GPT-5.4 mini (Medium)19.7237K-EXAONE 2.0 080319.7238Step 3.7 Flash19.5239Kimi K2.5 (Non-reasoning)19.4240Qwen3.5 27B (Non-reasoning)19.4241Claude 4.5 Sonnet (Non-reasoning)19.3242Qwen3.5 35B A3B (Reasoning)19.3243LongCat 2.019.1244Claude 4 Sonnet (Reasoning)18.9245DeepSeek V4 Flash 0420 (Non-reasoning)18.9246JT-35B-Flash18.7247MiniMax-M218.6248KAT-Coder-Pro V118.6249Claude 4.1 Opus (Non-reasoning)18.6250GLM-4.6 (Reasoning)18.5251GPT-6 Luna (Non-reasoning)18.5252Qwen3.5 397B A17B (Reasoning)18.4253MiMo-V2.5-Pro (Non-reasoning)18.3254Qwen3.6 35B A3B (Reasoning)18.2255GPT-5.4 (Non-reasoning)18.2256Grok 4 Fast (Reasoning)17.9257Gemini 3 Flash Preview (Non-reasoning)17.9258Qwen3.5 122B A10B (Non-reasoning)17.7259Claude 3.7 Sonnet (Reasoning)17.7260Muse Glimmer (High)17.5261GLM-4.7 (Non-reasoning)17.4262Hy3-preview (Non-reasoning)17.0263Ling-2.6-1T17.0264GPT-5.2 (Non-reasoning)17.0265Step 3.5 Flash 260317.0266Doubao Seed Code16.9267Claude 4.5 Haiku (Reasoning)16.9268GPT-5 mini (High)16.8269Gemma 4 26B A4B (Reasoning)16.7270o4-mini (High)16.7271Ring-2.6-1T16.6272Step 3.5 Flash16.6273Claude 4 Opus (Non-reasoning)16.6274Claude 4 Sonnet (Non-reasoning)16.6275DeepSeek V3.2 Exp (Reasoning)16.6276Qwen3 Max Thinking (Preview)16.3277Gemini 2.5 Pro16.1278MiMo-V2-Flash (Non-reasoning)16.0279DeepSeek V3.2 (Non-reasoning)16.0280K2 Horizon 3.7B15.6281Qwen3 Max15.6282Qwen3.5 122B A10B (Reasoning)15.6283Gemini 3.1 Flash-Lite15.6284GPT-5.6 Luna (Non-reasoning)15.5285Gemini 2.5 Flash Preview (Sep '25) (Reasoning)15.5286Claude 4.5 Haiku (Non-reasoning)15.4287Kimi K2 090515.3288Claude 3.7 Sonnet (Non-reasoning)15.3289Qwen3.6 35B A3B (Non-reasoning)15.2290o115.2291Qwen3.5 35B A3B (Non-reasoning)15.1292Gemini 2.5 Pro Preview (Mar' 25)15.0293GLM-4.6 (Non-reasoning)14.9294GLM-4.7-Flash (Reasoning)14.9295Granite 4.2 30B14.8296DeepSeek V3.1 Terminus (Reasoning)14.8297Gemma 4 31B (Reasoning)14.7298Grok 3 mini Reasoning (High)14.6299Grok 4.20 0309 (Non-reasoning)14.6300Gemini 2.5 Pro Preview (May' 25)14.5301DeepSeek V3.2 Speciale14.5302K-EXAONE (Reasoning)14.4303ERNIE 5.0 Thinking Preview14.3304Grok 4.20 0309 v2 (Non-reasoning)14.2305Mistral Medium 3.514.2306Gemma 4 12B (Reasoning)14.2307Nova 2.0 Pro Preview (Medium)14.2308Grok Code Fast 114.1309Grok 4.3 (Non-reasoning)14.0310DeepSeek V3.1 Terminus (Non-reasoning)13.9311Gemma 4 31B (Non-reasoning)13.9312DeepSeek V3.2 Exp (Non-reasoning)13.9313Apriel-v1.5-15B-Thinker13.8314Mercury 213.8315DeepSeek V3.1 (Non-reasoning)13.7316Nova 2.0 Omni (Medium)13.6317DeepSeek V3.1 (Reasoning)13.5318Qwen3 VL 235B A22B (Reasoning)13.4319Apriel-v1.6-15B-Thinker13.4320Nova 2.0 Lite (High)13.4321GPT-5.1 (Non-reasoning)13.3322Qwen3.5 9B (Non-reasoning)13.3323EXAONE 4.5 33B (Reasoning)13.2324Command A+13.1325Gemma 4 26B A4B (Non-reasoning)13.1326Qwen3.5 4B (Reasoning)13.1327DeepSeek R1 0528 (May '25)13.1328Gemini 2.5 Flash (Reasoning)13.1329GPT-5 nano (High)13.0330Nemotron 3.5 Lightning12.9331Nemotron 3 Super 120B A12B (Reasoning)12.8332Nova 2.0 Pro Preview (Low)12.8333GLM-4.5 (Reasoning)12.8334Kimi K212.7335Qwen3 235B A22B 2507 (Reasoning)12.7336GPT-4.112.7337Qwen3 Max (Preview)12.6338Nova 2.0 Lite (Medium)12.5339GPT-5 nano (Medium)12.5340Qwen3.5 Omni Flash12.5341o3-mini12.5342MiniCPM5-2B12.5343o1-pro12.4344Gemini 2.5 Flash Preview (Sep '25) (Non-reasoning)12.4345Mercury 2.512.3346JT-MINI12.2347Grok 312.1348Seed-OSS-36B-Instruct12.1349Qwen3 235B A22B 2507 Instruct12.0350Qwen3 Coder 480B A35B Instruct11.9351Qwen3 VL 32B (Reasoning)11.9352Magistral Medium 1.211.8353Sonar Reasoning Pro11.8354Nova 2.0 Lite (Low)11.8355HyperNova 60B 2605 (High, Based on gpt-oss-120b)11.7356MiniMax M1 80k11.7357GPT-5.4 nano (Non-reasoning)11.7358Nemotron Cascade 2 30B A3B11.7359Gemini 2.5 Flash Preview (Reasoning)11.7360gpt-oss-120b (High)11.6361K2 Think V211.5362LongCat Flash Lite11.5363GPT-5 (Minimal)11.4364DeepSeek R1 (Jan '25)11.4365o1-preview11.4366HyperCLOVA X SEED Think (32B)11.4367Grok 4.1 Fast (Non-reasoning)11.3368Mistral Small 4 (Reasoning)11.3369GLM-4.6V (Reasoning)11.2370K-EXAONE (Non-reasoning)11.2371Qwen3 Next 80B A3B (Reasoning)11.2372Qwen3.5 9B (Reasoning)11.2373GPT-5.4 mini (Non-reasoning)11.1374Nova 2.0 Omni (Low)11.1375GLM-4.5-Air11.1376Granite 4.2 8B11.1377Grok 4 Fast (Non-reasoning)11.1378Ling 3.0 Tiny11.1379Mi:dm K 2.5 Pro11.0380o3-mini (High)11.0381Ring-1T10.9382G9v3-3B10.8383Trinity Large Thinking10.8384Qwen3.5 4B (Non-reasoning)10.8385INTELLECT-3 (Based on GLM-4.5-Air)10.6386GLM-4.7-Flash (Non-reasoning)10.6387GPT-5 (ChatGPT)10.4388Solar Open 100B (Reasoning)10.4389Grok 3 Reasoning Beta10.4390Gemini 2.5 Flash-Lite Preview (Sep '25) (Reasoning)10.4391Nemotron 3 Nano Omni 30B A3B Reasoning10.3392gpt-oss-120b (Low)10.2393GPT-4.1 mini10.2394Llama 4 Maverick10.0395MiniMax M1 40k10.0396Nova 2.0 Pro Preview (Non-reasoning)10.0397gpt-oss-20b (Low)10.0398Qwen3 VL 235B A22B Instruct9.9399North Mini Code9.9400GPT-5 mini (Minimal)9.9401K2-V2 (High)9.9402Gemini 2.5 Flash (Non-reasoning)9.9403Qwen3 30B A3B 2507 (Reasoning)9.8404o1-mini9.8405DeepSeek V3 03249.7406Ling 2.6 Flash9.7407Qwen3 Next 80B A3B Instruct9.6408Tri-21B-think Preview9.6409Qwen3 Coder 30B A3B Instruct9.6410GPT-4.5 (Preview)9.6411DiffusionGemma 26B A4B9.5412Qwen3 235B A22B (Reasoning)9.5413QwQ 32B9.5414Qwen3 VL 30B A3B (Reasoning)9.5415Gemini 2.0 Flash Thinking Experimental (Jan '25)9.4416Gemma 4 12B (Non-reasoning)9.4417Gemini 2.5 Flash-Lite Preview (Sep '25) (Non-reasoning)9.3418Mistral Large 39.3419Qwen3 Coder Next9.2420Motif-2-12.7B-Reasoning9.2421Mistral Medium 3.19.2422Ling-1T9.2423Nova Premier9.2424Solar Pro 2 (Preview) (Reasoning)9.1425Granite 4.2 3B9.1426Magistral Medium 19.1427Mistral Medium 39.0428K2-V2 (Medium)9.0429Llama Nemotron Super 49B v1.5 (Reasoning)9.0430Devstral Medium9.0431Mistral Small 4 (Non-reasoning)9.0432Tri-21B-Think9.0433gpt-oss-20b (High)9.0434GPT-4o (March 2025, chatgpt-4o-latest)9.0435Gemini 2.0 Flash (Feb '25)8.9436Claude 3.5 Haiku8.9437Llama 3.3 Nemotron Super 49B v1 (Reasoning)8.9438Gemma 4 E4B (Reasoning)8.9439NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)8.9440Qwen3 4B 2507 (Reasoning)8.8441MiniCPM5-1B (Reasoning)8.8442Sarvam 105B (High)8.8443Gemini 2.0 Pro Experimental (Feb '25)8.7444Nova 2.0 Lite (Non-reasoning)8.7445Devstral Small (May '25)8.7446Claude 3 Opus8.7447MiniCPM5-1B (Non-reasoning)8.7448Sonar Reasoning8.7449Gemini 2.5 Flash Preview (Non-reasoning)8.7450Devstral 28.6451Magistral Small 1.28.6452Qwen3 32B (Reasoning)8.6453Gemini 2.5 Flash-Lite (Reasoning)8.5454DeepSeek V3 (Dec '24)8.5455GPT-4o (Nov '24)8.4456Nanbeige4.1-3B8.4457LFM2.5-2.6B8.4458Qwen3 VL 32B Instruct8.4459DeepSeek R1 Distill Qwen 32B8.4460GLM-4.6V (Non-reasoning)8.4461Qwen3 235B A22B (Non-reasoning)8.3462Mistral Small 3.28.2463Magistral Small 18.2464Gemini 2.0 Flash (experimental)8.2465EXAONE 4.0 32B (Reasoning)8.2466Qwen3 VL 8B (Reasoning)8.2467Qwen3 14B (Reasoning)8.2468Nova 2.0 Omni (Non-reasoning)8.2469DeepSeek R1 0528 Qwen3 8B8.1470Llama 4 Scout8.1471Qwen2.5 Max8.0472Qwen3 VL 30B A3B Instruct7.9473Hermes 4 - Llama-3.1 70B (Reasoning)7.9474Gemini 1.5 Pro (Sep '24)7.9475Solar Pro 2 (Preview) (Non-reasoning)7.9476DeepSeek R1 Distill Llama 70B7.9477Claude 3.5 Sonnet (Oct '24)7.9478DeepSeek R1 Distill Qwen 14B7.8479Falcon-H1R-7B7.8480GPT-4.1 nano7.8481Solar Pro 37.8482Ling-flash-2.07.8483Gemma 4 E2B (Reasoning)7.8484Qwen3 Omni 30B A3B (Reasoning)7.8485GPT-4o (Aug '24)7.7486Qwen2.5 Instruct 72B7.7487Sonar7.7488Step3 VL 10B7.7489Llama 3.3 Instruct 70B7.7490Qwen3 30B A3B (Reasoning)7.6491Sonar Pro7.6492Devstral Small (Jul '25)7.6493QwQ 32B-Preview7.6494GLM-4.5V (Reasoning)7.6495Mistral Large 2 (Nov '24)7.6496Devstral Small 27.5497Llama 3.1 Nemotron Ultra 253B v1 (Reasoning)7.5498Qwen3 30B A3B 2507 Instruct7.5499ERNIE 4.5 300B A47B7.5500Hermes 4 - Llama-3.1 405B (Reasoning)7.5501Solar Pro 2 (Reasoning)7.5502NVIDIA Nemotron Nano 12B v2 VL (Reasoning)7.5503Gemma 4 E4B (Non-reasoning)7.5504Granite 4.1 30B7.4505NVIDIA Nemotron Nano 9B V2 (Reasoning)7.4506Hermes 4 - Llama-3.1 405B (Non-reasoning)7.4507Gemini 2.0 Flash-Lite (Feb '25)7.4508Llama Nemotron Super 49B v1.5 (Non-reasoning)7.4509Qwen3 32B (Non-reasoning)7.3510GPT-4o (May '24)7.3511Gemini 2.0 Flash-Lite (Preview)7.3512K2-V2 (Low)7.3513Llama 3.1 Nemotron Nano 4B v1.1 (Reasoning)7.3514Kimi Linear 48B A3B Instruct7.3515Llama 3.1 Instruct 405B7.3516Qwen3 8B (Reasoning)7.3517Llama 3.3 Nemotron Super 49B v1 (Non-reasoning)7.3518Qwen3 VL 8B Instruct7.3519Qwen3 4B (Reasoning)7.2520LFM2.5-8B-A1B7.2521Claude 3.5 Sonnet (June '24)7.2522Llama 3.1 Tulu3 405B7.2523GPT-4o (ChatGPT)7.2524Ring-flash-2.07.2525Pixtral Large7.1526Olmo 3.1 32B Think7.1527Mistral Small 3.17.1528Grok 2 (Dec '24)7.1529GPT-5 nano (Minimal)7.1530Gemini 1.5 Flash (Sep '24)7.1531Qwen3 VL 4B (Reasoning)7.0532GPT-4 Turbo7.0533Solar Pro 2 (Non-reasoning)7.0534Nova Pro7.0535Command A7.0536Qwen3.5 2B (Reasoning)6.9537Llama 3.1 Nemotron Instruct 70B6.9538Llama 3.1 Instruct 8B6.9539Grok Beta6.9540Qwen2.5 Instruct 32B6.9541NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)6.8542NVIDIA Nemotron Nano 9B V2 (Non-reasoning)6.8543Mistral Large 2 (Jul '24)6.8544Qwen3 4B 2507 Instruct6.7545Qwen2.5 Coder Instruct 32B6.7546Qwen3 14B (Non-reasoning)6.7547GPT-46.7548GLM-4.5V (Non-reasoning)6.7549Mistral Small 36.7550Gemini 2.5 Flash-Lite (Non-reasoning)6.7551Nova Lite6.7552GPT-4o mini6.7553Hermes 4 - Llama-3.1 70B (Non-reasoning)6.7554Qwen3 30B A3B (Non-reasoning)6.6555DeepSeek-V2.5 (Dec '24)6.6556Qwen3 4B (Non-reasoning)6.6557Llama 3.1 Instruct 70B6.6558Granite 4.1 8B6.6559Sarvam 30B (High)6.6560Gemini 2.0 Flash Thinking Experimental (Dec '24)6.6561DeepSeek-V2.56.6562Olmo 3.1 32B Instruct6.5563Mistral Saba6.5564DeepSeek R1 Distill Llama 8B6.5565Gemma 4 E2B (Non-reasoning)6.5566Olmo 3 32B Think6.5567Gemini 1.5 Pro (May '24)6.4568R1 17766.4569Qwen2.5 Turbo6.4570Reka Flash (Sep '24)6.4571Llama 3.2 Instruct 90B (Vision)6.4572Solar Mini6.4573Celeris-16.3574Grok-16.3575Qwen2 Instruct 72B6.3576Phi-4 Mini Instruct6.3577EXAONE 4.0 32B (Non-reasoning)6.3578Qwen3.5 2B (Non-reasoning)6.2579Gemini 1.5 Flash-8B6.2580Qwen3.5 0.8B (Reasoning)6.1581DeepHermes 3 - Mistral 24B Preview (Non-reasoning)6.1582Jamba 1.7 Large6.1583Granite 4.0 H Small6.0584Ministral 3 14B6.0585Jamba 1.5 Large6.0586Qwen3 Omni 30B A3B Instruct6.0587Hermes 3 - Llama-3.1 70B6.0588Qwen3 8B (Non-reasoning)6.0589DeepSeek-Coder-V26.0590OLMo 2 32B6.0591Jamba 1.6 Large6.0592LFM2 24B A2B5.9593Gemini 1.5 Flash (May '24)5.9594Phi-45.9595Claude 3 Sonnet5.9596Nova Micro5.9597Granite 4.1 3B5.9598Mistral Small (Sep '24)5.8599Gemini 1.0 Ultra5.8600Phi-3 Mini Instruct 3.8B5.8601NVIDIA Nemotron Nano 12B v2 VL (Non-reasoning)5.8602Gemma 3n E4B Instruct Preview (May '25)5.8603Phi-4 Multimodal Instruct5.8604Qwen2.5 Coder Instruct 7B5.8605Mistral Large (Feb '24)5.8606Mixtral 8x22B Instruct5.7607Llama 2 Chat 7B5.7608Llama 3.2 Instruct 3B5.7609MiniCPM-V 4.6 1.3B5.7610Jamba Reasoning 3B5.7611Qwen3 VL 4B Instruct5.7612Qwen1.5 Chat 110B5.7613Reka Flash 35.6614Olmo 3 7B Think5.6615Claude 2.15.6616Claude 3 Haiku5.6617OLMo 2 7B5.6618Molmo 7B-D5.6619Ling-mini-2.05.5620DeepSeek R1 Distill Qwen 1.5B5.5621Claude 2.05.5622DeepSeek-V2-Chat5.5623Mistral Small (Feb '24)5.5624Mistral Medium5.5625GPT-3.5 Turbo5.5626Ministral 3 8B5.5627Llama 3 Instruct 70B5.5628Arctic Instruct5.4629Qwen Chat 72B5.4630LFM 40B5.4631Llama 3.2 Instruct 11B (Vision)5.4632Qwen3.5 0.8B (Non-reasoning)5.4633PALM-25.4634Gemini 1.0 Pro5.3635DeepSeek Coder V2 Lite Instruct5.3636Sarvam M (Reasoning, Based on Mistral Small 3.1)5.3637DeepSeek LLM 67B Chat (V1)5.3638Llama 2 Chat 70B5.3639Llama 2 Chat 13B5.3640Command-R+ (Apr '24)5.3641OpenChat 3.5 (1210)5.3642DBRX Instruct5.3643Exaone 4.0 1.2B (Reasoning)5.3644Olmo 3 7B Instruct5.2645Exaone 4.0 1.2B (Non-reasoning)5.2646LFM2.5-1.2B-Thinking5.2647Jamba 1.7 Mini5.2648LFM2 2.6B5.2649LFM2.5-1.2B-Instruct5.2650Jamba 1.5 Mini5.2651Granite 4.0 H 1B5.2652Qwen3 1.7B (Reasoning)5.2653Jamba 1.6 Mini5.2654Mixtral 8x7B Instruct5.1655Gemma 3 270M5.1656Apertus 70B Instruct5.1657Granite 4.0 Micro5.1658DeepHermes 3 - Llama-3.1 8B Preview (Non-reasoning)5.1659Claude Instant5.0660Command-R (Mar '24)5.0661Llama 65B5.0662Mistral 7B Instruct5.0663Qwen Chat 14B5.0664Granite 4.0 1B5.0665Molmo2-8B5.0666LFM2 8B A1B4.9667Granite 3.3 8B (Non-reasoning)4.9668Qwen3 1.7B (Non-reasoning)4.9669Gemma 3 27B Instruct4.9670Ministral 3 3B4.8671Apertus 8B Instruct4.8672Gemma 3 1B Instruct4.8673Gemma 3 4B Instruct4.8674Gemma 3n E2B Instruct4.8675Gemma 3n E4B Instruct4.8676Granite 4.0 350M4.8677Granite 4.0 H 350M4.8678LFM2 1.2B4.8679LFM2.5-VL-1.6B4.8680Llama 3 Instruct 8B4.8681Llama 3.2 Instruct 1B4.8682Qwen3 0.6B (Non-reasoning)4.8683Qwen3 0.6B (Reasoning)4.8684Tiny Aya Global4.8685NVIDIA Nemotron 3 Nano 4B4.8686Gemma 3 12B Instruct3.8687K2 Horizon 0.9B3.0数值越大越好 · 可上下滚动查看更多
智能指数 vs 价格
性价比缩放 141%
横轴：综合价格（$/1M，对数坐标）· 纵轴：智能指数 · 滚轮以屏幕中心等比缩放（带缓动）
悬停查看详情，点击跳转模型页 · 滚轮缩放、拖动平移 · 虚线为拟合趋势线（对数刻度下呈直线，等差刻度下呈曲线）
默认按发布日期倒序排序，点击任意列表头可切换排序 · 同一型号只保留最强等级
| # | 模型 | 供应商 | 发布日期↓ | 价格-输入 | 价格-输出 | 输出速度 | 首token时间 | 首答案token时间 | 智能指数 | 代码指数 | 数学指数 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Claude Haiku 5.5 (Max) | Anthropic | 2026-10-07 | ¥0.67 | ¥3.36 | 239 | 277.59s | 277.59s | 43.40 | - | - |
| 2 | Mistral Large 4 Preview | Mistral | 2026-10-06 | ¥9.13 | ¥28.07 | 106 | 0.97s | 19.82s | 38.40 | - | - |
| 3 | Ling 3.1 Flash | InclusionAI | 2026-10-01 | ¥2.01 | ¥6.04 | 216 | 1.12s | 10.39s | 41.10 | - | - |
| 4 | Gemini 4 Argon (High) | Google | 2026-09-30 | ¥13.43 | ¥67.15 | - | - | - | 52.60 | - | - |
| 5 | GPT-6.1 Sol (Max) | OpenAI | 2026-09-29 | ¥13.43 | ¥67.15 | 58 | 176.52s | 176.52s | 51.80 | - | - |
| 6 | Claude Sonnet 5.5 (Max, Default Fallback) | Anthropic | 2026-09-28 | ¥13.43 | ¥67.15 | 134 | 270.90s | 270.90s | 56.00 | - | - |
| 7 | Claude Opus 5.5 (Max, Default Fallback) | Anthropic | 2026-09-22 | ¥26.86 | ¥134.31 | 98 | 560.79s | 560.79s | 57.60 | - | - |
| 8 | GPT-6 Sol (Max) | OpenAI | 2026-09-22 | ¥13.43 | ¥67.15 | - | - | - | 47.60 | - | - |
| 9 | GPT-6 Luna (Max) | OpenAI | 2026-09-22 | ¥0.67 | ¥3.36 | 129 | 92.49s | 92.49s | 38.10 | - | - |
| 10 | Solar Mini 4 | Upstage | 2026-09-22 | ¥0.67 | ¥2.69 | 65 | 1.24s | 32.09s | 24.10 | - | - |
| 11 | Grok 4.7 (Xhigh) | SpaceXAI | 2026-09-21 | ¥13.43 | ¥40.29 | 68 | 14.04s | 14.04s | 46.40 | - | - |
| 12 | MiMo-V2.6-Pro | Xiaomi | 2026-09-21 | ¥2.92 | ¥5.84 | 38 | 4.16s | 56.58s | 46.30 | - | - |
| 13 | MiMo-V2.6-Flash | Xiaomi | 2026-09-21 | ¥0.94 | ¥1.88 | 57 | 4.26s | 39.48s | 37.90 | - | - |
| 14 | Step 5 Preview | StepFun | 2026-09-18 | ¥6.72 | ¥18.13 | 90 | 1.86s | 24.16s | 43.70 | - | - |
| 15 | Ling-3.0-flash-Fin | InclusionAI | 2026-09-11 | ¥0.50 | ¥1.48 | 332 | 1.25s | 7.27s | 22.60 | 55.60 | - |
| 16 | DeepSeek V4.1 Flash (Max) | DeepSeek | 2026-09-10 | ¥2.01 | ¥8.06 | 216 | 0.86s | 10.10s | 39.50 | - | - |
| 17 | Ling-3.0-flash-VL | InclusionAI | 2026-09-10 | ¥0.50 | ¥1.48 | 149 | 1.32s | 14.75s | 24.60 | 57.00 | - |
| 18 | Mercury 2.5 | Inception | 2026-09-08 | ¥1.68 | ¥5.04 | 697 | 2.75s | 2.75s | 12.30 | - | - |
| 19 | MiniCPM5-2B | OpenBMB | 2026-09-07 | - | - | - | - | - | 12.50 | 14.50 | - |
| 20 | GPT-6 Astra (Max) | OpenAI | 2026-09-03 | ¥67.15 | ¥335.76 | 47 | 235.74s | 235.74s | 52.70 | 76.90 | - |
| 21 | K2 Horizon 375B A23B | Institute of Foundation Models | 2026-09-03 | - | - | 129 | 2.12s | 17.64s | 30.50 | 61.50 | - |
| 22 | K2 Horizon MoVA 36B A4B | Institute of Foundation Models | 2026-09-03 | - | - | - | - | - | 25.30 | - | - |
| 23 | K2 Horizon 7B | Institute of Foundation Models | 2026-09-03 | - | - | - | - | - | 20.60 | 38.60 | - |
| 24 | K2 Horizon 3.7B | Institute of Foundation Models | 2026-09-03 | - | - | - | - | - | 15.60 | 26.10 | - |
| 25 | K2 Horizon 0.9B | Institute of Foundation Models | 2026-09-03 | - | - | - | - | - | 3.00 | 3.40 | - |
| 26 | Muse Spark 1.3 (Max) | Meta | 2026-09-02 | ¥8.39 | ¥28.54 | 174 | 41.22s | 52.70s | 48.10 | 75.80 | - |
| 27 | Qwen3.8 Max (0902) | Alibaba | 2026-09-02 | ¥13.43 | ¥40.29 | 37 | 1.77s | 56.26s | 45.40 | 76.20 | - |
| 28 | Gemini 3.8 Flash (High) | Google | 2026-09-02 | ¥5.04 | ¥25.18 | 131 | 17.98s | 17.98s | 40.90 | 76.30 | - |
| 29 | Claude Fable 5.1 (Max, Default Fallback) | Anthropic | 2026-09-01 | ¥67.15 | ¥335.76 | 68 | 149.81s | 149.81s | 53.40 | 81.60 | - |
| 30 | Apodex 1.1 | Apodex | 2026-08-30 | ¥2.01 | ¥20.15 | - | - | - | 26.40 | 60.80 | - |
| 31 | GLM-5.3-Flash | Z AI | 2026-08-26 | ¥1.01 | ¥3.36 | 51 | 2.81s | 42.26s | 41.80 | 71.50 | - |
| 32 | Qwen3.8-Flash-Next | Alibaba | 2026-08-26 | ¥1.01 | ¥3.16 | 54 | 1.55s | 38.72s | 39.80 | 73.10 | - |
| 33 | Granite 4.2 30B | IBM | 2026-08-25 | ¥1.07 | ¥4.36 | 77 | 0.29s | 26.21s | 14.80 | 29.90 | - |
| 34 | Granite 4.2 8B | IBM | 2026-08-25 | ¥0.40 | ¥1.68 | 53 | 0.37s | 38.29s | 11.10 | 22.40 | - |
| 35 | Granite 4.2 3B | IBM | 2026-08-25 | ¥0.20 | ¥0.81 | 226 | 0.23s | 9.06s | 9.10 | 17.50 | - |
| 36 | DeepSeek V4 Flash Vision (Max) | DeepSeek | 2026-08-21 | ¥2.95 | ¥8.86 | 214 | 0.87s | 10.23s | 34.80 | 65.00 | - |
| 37 | G9v3-39A5B | AI9Stars | 2026-08-20 | - | - | - | - | - | 21.80 | 33.10 | - |
| 38 | GLM-5.3 (Max) | Z AI | 2026-08-18 | ¥9.40 | ¥29.55 | 68 | 2.75s | 32.22s | 44.80 | 74.80 | - |
| 39 | Qwen3.8 27B (Xhigh) | Alibaba | 2026-08-14 | ¥3.36 | ¥20.15 | 47 | 1.18s | 43.89s | 33.70 | 68.10 | - |
| 40 | Gemini 3.7 Flash (Medium) | Google | 2026-08-13 | ¥5.04 | ¥25.18 | - | - | - | 39.60 | 71.50 | - |
共显示 40 / 450 个模型 · 数据来源：Artificial Analysis 公开评测
