跳过主要内容
参与排名推理

CritPt

Artificial Analysis · CritPt 的公开模型评测结果。

官方评测
来源说明

该来源测量模型在「推理」领域的具体交付表现,原始成绩会和统一能力读数分开展示。

限制: 不同版本、推理设置和测试集可能导致分数变化,请结合数据时间阅读。

使用方式

参与排名的来源会影响能力分;参考、交叉参考和观察来源仅用于对比与核验。

本来源的模型成绩

Muse Wonder Full0.3285714387.32026-10-18
Muse Wonder Full0.3285714387.32026-11-05
Muse Wonder Full0.3285714387.32026-10-17
GPT-5.6 Sol0.3228571486.52026-11-05
GPT-6 Astra0.3171428685.72026-11-05
Claude Opus 5.50.3171428685.72026-11-05
GPT-6.1 Sol0.3171428685.72026-11-05
Claude Sonnet 5.50.3142857185.22026-11-05
Claude Fable 5.10.3114285784.82026-11-05
GPT-6 Sol0.3085714384.42026-11-05
GPT-5.5 Pro0.3057142984.02026-11-05
GPT-5.4 Pro0.383.12026-11-05
GPT-5.6 Terra0.383.12026-11-05
Claude Opus 50.2914285781.92026-11-05
Claude Fable 50.2857142981.02026-11-05
Muse Wonder0.2828571480.62026-10-18
Muse Wonder0.2828571480.62026-10-17
Muse Wonder0.2828571480.62026-11-05
GPT-6.1 Terra0.2880.22026-11-05
GPT-5.50.2714285778.92026-11-05
Gemini 4 Argon0.2714285778.92026-11-05
MiMo V2.6 Pro0.2657142978.12026-11-05
Ling-3.1-flash-10190.2677.22026-11-05
Muse Spark 1.30.2677.22026-11-05
DeepSeek V4.1 Flash 10170.2474.32026-11-05
GPT-5.40.2342857173.42026-11-05
Kimi K30.2342857173.42026-11-05
Claude Haiku 5.50.2257142972.22026-11-05
MiniMax M3.1 Flash0.2228571471.72026-11-05
Claude Opus 4.80.2085714369.62026-11-05
GLM-5.20.2085714369.62026-11-05
Step 5 Preview0.2085714369.62026-11-05
GPT-5.6 Luna0.2057142969.22026-11-05
Qwen3.8 Max0.268.42026-11-05
Grok 4.60.1971428667.92026-11-05
GPT-6 Luna0.1942857167.52026-11-05
GLM 5.30.1914285767.12026-11-05
Gemini 3.8 Flash0.1828571465.82026-11-05
Ling-3.1-flash0.1865.42026-11-05
Ling-3.1-flash0.1865.42026-10-17
Ling-3.1-flash0.1865.42026-10-18
DeepSeek V4 Pro 08130.1865.42026-11-05
Grok 4.70.1865.42026-11-05
Gemini 3.1 Pro Preview0.1771428665.02026-11-05
Muse Spark 1.20.1771428665.02026-11-05
Claude Sonnet 50.1685714363.72026-11-05
DeepSeek V4 Flash 07310.1657142963.32026-11-05
Grok 4.50.1542857161.62026-11-05
GLM-5.3 Flash0.1542857161.62026-10-17
GLM-5.3 Flash0.1542857161.62026-10-18
GLM-5.3 Flash0.1542857161.62026-11-05
Muse Spark 1.10.1514285761.22026-11-05
DeepSeek V4.1 Flash0.1428571459.92026-11-05
Qwen3.7 Max0.1342857158.72026-11-05
DeepSeek V4 Pro0.1285714357.82026-11-05
Claude Opus 4.60.1257142957.42026-11-05
MiMo V2.6 Flash0.1256.62026-10-17
MiMo V2.6 Flash0.1256.62026-10-18
MiMo V2.6 Flash0.1256.62026-11-05
Claude Opus 4.70.1256.62026-11-05
GPT-5.20.115510255.92026-11-05
Mistral Large 40.1114285755.32026-11-05
Gemini 3.6 Flash0.1057142954.52026-11-05
Mistral Large 40.1057142954.52026-10-28
GPT-5.4 mini0.153.62026-11-05
Kimi K2.7 Code0.153.62026-11-05
Gemini 3 Pro Preview0.0914285752.32026-11-05
Qwen3.7 Plus0.0914285752.32026-11-05
Gemini 3 Flash Preview0.0857142951.52026-11-05
Nex-N2-Pro0.0857142951.52026-11-05
Inkling Small0.0828571451.12026-11-05
Grok 4.30.0850.72026-11-05
Kimi K2.60.0850.72026-11-05
Grok 4.200.0657142948.52026-10-11
Grok 4.20 (Reasoning)0.0647.72026-11-05
GPT-50.0571428647.32026-11-05
Inkling0.0542857146.92026-11-05
Hy30.0485714346.02026-11-05
Claude Opus 4.50.0457142945.62026-11-05
GLM-5.10.0457142945.62026-11-05
Hy3 preview0.0457142945.62026-11-05
MiMo-V2-Flash0.0428571445.22026-11-05
MiMo-V2.5-Pro0.0444.82026-11-05
MiMo-V2.50.0371428644.32026-11-05
Ring-2.6-1T0.0371428644.32026-11-05
MiniMax-M30.0371428644.32026-11-05
Qwen3.6 Max Preview0.0370612244.32026-11-05
Kimi K2.50.0314285743.52026-11-05
Claude Sonnet 4.60.0314285743.52026-11-05
Nemotron 3 Super 120B A12B0.0314285743.52026-11-05
Nemotron 3 Ultra 550B A55B0.0314285743.52026-11-05
DeepSeek V3.20.0285714343.12026-11-05
Qwen3.6 Plus0.0285714343.12026-11-05
Gemini 2.5 Pro0.0257142942.62026-11-05
LongCat-2.00.0257142942.62026-11-05
Step 3.5 Flash0.0246938842.52026-11-05
Step 3.5 Flash 26030.0228571442.22026-11-05
Step 3.7 Flash0.0228571442.22026-11-05
DeepSeek V3.10.0241.82026-11-05
GLM-50.0241.82026-11-05
DeepSeek V3.1 Terminus0.0171428641.42026-11-05
Qwen3 Max Thinking0.0171428641.42026-11-05
Ling-3.0-flash (free)0.0171428641.42026-11-05
GLM-4.70.0171020441.42026-11-05
Qwen3.5 397B-A17B0.0170612241.42026-11-05
Gemini 2.5 Flash0.0142857141.02026-11-05
gpt-oss-20b0.0142857141.02026-11-05
GPT-5 mini0.0142857141.02026-11-05
DeepSeek V3.2 Exp0.0142857141.02026-11-05
DeepSeek V3.2 Exp0.0142857141.02026-10-27
o30.0114285740.52026-11-05
gpt-oss-120b0.0114285740.52026-11-05
Claude Sonnet 4.50.0114285740.52026-11-05
GLM-4.60.0114285740.52026-11-05
MiniMax-M2.50.0114285740.52026-11-05
MiMo-V2-Omni0.0114285740.52026-11-05
Qwen3.6 27B0.0114285740.52026-11-05
Gemini 3.1 Flash Lite0.0114285740.52026-11-05
MiniMax-M20.0085714340.12026-11-05
Nemotron 3 Nano 30B A3B0.0085714340.12026-11-05
Qwen3.5 122B-A10B0.0085714340.12026-11-05
Qwen3.5 35B-A3B0.0085714340.12026-11-05
Trinity Large Thinking0.0085714340.12026-11-05
Qwen3.5 27B0.0085306140.12026-11-05
Mercury 20.008489840.12026-11-05
DeepSeek-R10.0057142939.72026-11-05
o4-mini0.0057142939.72026-10-11
o4 Mini High0.0057142939.72026-11-05
Qwen3.5 9B0.0057142939.72026-11-05
MiniMax-M2.70.0057142939.72026-11-05
Nemotron Cascade 2 30B A3B0.0057142939.72026-11-05
GLM-5V-Turbo0.0057142939.72026-11-05
o10.0028571439.32026-11-03
o3-mini0.0028571439.32026-10-11
o3 Mini High0.0028571439.32026-11-05
Qwen3 32B0.0028571439.32026-11-05
MiniMax-M2.10.0028571439.32026-11-05
GLM-4.7-Flash0.0028571439.32026-11-05
Sarvam 30B0.0028571439.32026-11-05
GLM-5-Turbo0.0028571439.32026-11-05
MiMo-V2-Pro0.0028571439.32026-11-05
Qwen3.6 35B-A3B0.0028571439.32026-11-05
Ling-2.6-1T0.0028571439.32026-11-05
Command A Plus0.0028571439.32026-11-05
North Mini Code0.0028571439.32026-11-05
Gemma 4 12B Base038.82026-11-05
Qwen2.5 72B Instruct038.82026-11-05
Gemini 2.0 Flash038.82026-11-05
Phi 4038.82026-11-05
R1 Distill Llama 70B038.82026-11-05
Reka Flash 3038.82026-11-05
Command A038.82026-11-05
DeepSeek V3 0324038.82026-11-05
Qwen3 235B-A22B038.82026-11-05
Qwen3 235B-A22B038.82026-10-27
Qwen3-Coder 30B-A3B Instruct038.82026-11-05
Qwen3-Coder 480B-A35B Instruct038.82026-11-05
Llama 4 Maverick038.82026-11-05
Llama 4 Scout038.82026-11-05
Llama 3.3 Nemotron Super 49B v1038.82026-11-05
Qwen3 30B A3B038.82026-11-05
Qwen3 8B038.82026-11-05
Qwen3 8B038.82026-10-27
Qwen3 14B038.82026-11-05
Qwen3 14B038.82026-10-27
Gemini 2.5 Flash-Lite038.82026-11-05
Gemini 2.5 Flash-Lite038.82026-10-27
Qwen3 235B A22B Instruct 2507038.82026-11-05
GLM-4.5038.82026-11-05
GLM-4.5-Air038.82026-11-05
Qwen3 30B A3B Instruct 2507038.82026-11-05
GPT-5 nano038.82026-11-05
GPT-5 nano038.82026-10-27
GLM-4.5V038.82026-11-05
GLM-4.5V038.82026-10-27
Nemotron Nano 9B v2038.82026-11-05
Nemotron Nano 9B v2038.82026-10-27
Qwen3-Next 80B-A3B Instruct038.82026-11-05
Sarvam 105B038.82026-11-05
Kimi K2 0905038.82026-11-05
Qwen3 Max038.82026-11-05
Qwen3 VL 235B A22B Instruct038.82026-11-05
Qwen3 VL 30B A3B Instruct038.82026-11-05
Qwen3 VL 8B Instruct038.82026-11-05
Claude Haiku 4.5038.82026-11-05
Granite 4.0 Micro038.82026-11-05
Qwen3 VL 32B Instruct038.82026-11-05
Nemotron Nano 12B v2 VL038.82026-11-05
Olmo 3 32B Think038.82026-11-05
Mistral Large 3038.82026-11-05
Ministral 3 14B038.82026-11-05
Ministral 3 8B038.82026-11-05
Ministral 3 3B038.82026-11-05
GLM-4.6V038.82026-11-05
GLM-4.6V038.82026-10-27
Solar Pro 3038.82026-11-05
Qwen3 Coder Next038.82026-11-05
Grok 4.20 (Non-Reasoning)038.82026-11-05
KAT-Coder-Pro V2038.82026-11-05
Ling-2.6-flash038.82026-11-05
Nemotron 3 Nano Omni 30B A3B Reasoning038.82026-11-05
Mistral Medium 3.5038.82026-11-05
Mistral Medium 3.5038.82026-11-05
Granite 4.1 8B038.82026-11-05
Gemini 3.5 Flash Lite038.82026-11-05

常见问题