VECTOR WIREAI INTELLIGENCE
NVDA$1,847+3.2%MSFT$512+1.1%GOOGL$199-0.4%META$728+2.7%AMD$184-1.2%TSM$212+0.6%PLTR$98+4.1%AI IDX4,821+1.9%
PKT
SEEDRefresh Models Deals Regulatory Sources

GPQA (unspecified)

31 models tracked

Review bucket: GPQA rows whose raw labels cannot prove the split (Diamond vs main). Rows leave ONLY on a stated evidence rule (e.g. score matches the model's verified Diamond ≤2pts). Never pooled.

#ModelVendorBest scoreRunsLast seen
1GPT-5.2OpenAI92.412026-05-01
2Kimi K2.6 1T A32BMoonshot9112026-06-06
3Claude Sonnet 4.6Anthropic89.912026-05-01
4Qwen 3.5 397B 17BAlibaba87.112026-06-06
5MiniMax 2.7 230B A10BMiniMax86.612026-06-06
6GLM-5.1-744B-A40BZ.ai86.112026-06-06
7Qwen3.6 35B A3BAlibaba8612026-05-03
8GPT-5OpenAI85.712026-05-15
9Gemma 4 31BGoogle84.312026-06-06
10Qwen3.5 35B A3BAlibaba84.212026-06-06
11Claude Sonnet 4.5Anthropic83.422026-06-12
12Gemma 4 26B A4BGoogle82.312026-06-06
13DeepSeek-V3.2DeepSeek79.912026-05-15
14GLM-4.7-FlashZ.ai75.212026-05-16
15Kimi K2 (Reasoning)Moonshot74.212026-05-15
16Qwen3 30B A3BAlibaba73.412026-05-16
17gpt-oss-20bOpenAI71.512026-05-16
18Qwen2.5 7B BaseAlibaba63.712026-05-03
19Qwen2.5 3B BaseAlibaba56.912026-05-03
20Nemotron 3 Ultra 550B A55B BaseNVIDIA5012026-06-05
21Granite 4.1 30BIBM45.812026-06-15
22MiMo V2 Flash BaseXiaomi43.522026-06-05
23Kimi K2 BaseMoonshot43.432026-06-05
24DeepSeek-V3.1-BaseDeepSeek43.122026-06-05
25Qwen 2.5 14BAlibaba42.912026-05-30
26Step3.5 Flash BaseStepFun41.722026-06-05
27DeepSeek-V3.2-Exp-BaseDeepSeek37.332026-06-05
28Mistral Large 3 675B Base 2512Mistral34.912026-06-05
29GLM-4.5-BaseZ.ai34.932026-06-05
30Phi 3 Medium 4K InstructMicrosoft31.212026-05-30
31Gemma 3 1B Base DenseGoogle2612026-05-03

Best tracked score per model (default configuration; source-attributed and verification-tiered). Open a model for its full benchmark surface, provenance and pricing.