Skip to content

AI model leaderboard

Factual Q&A and graduate-level scientific knowledge, testing knowledge mastery and answer accuracy.

2 evaluations · 1 organisations · updated 10/01 14:05

Knowledgetop 30

Ranked on the evidence several public evaluations share
The top 30 models, sortable by column
Model
01GPT-6 AstraOpenAIReleased 2026-09-03 · Evidence-sensitiveCached input ¥6.7Input ¥67.05 · Output ¥335.2396.9
02gpt-6.1-sol—Released — · Evidence-sensitiveCached input —Input — · Output —95.4
03Gemini 3.1 Pro PreviewGoogleReleased 2026-02-19 · Evidence-sensitiveCached input ¥1.34Input ¥13.41 · Output ¥80.4594.5
04Claude Opus 5.5AnthropicReleased 2026-09-17 · Evidence-sensitiveCached input ¥1.34Input ¥26.82 · Output ¥134.0991.4
05Gemini 3.8 FlashGoogleReleased 2026-09-02 · Evidence-sensitiveCached input ¥0.5Input ¥5.03 · Output ¥25.1490.1
06Gemini 3.7 FlashGoogleReleased 2026-08-13 · Evidence-sensitiveCached input ¥0.5Input ¥5.03 · Output ¥25.1488.8
07GPT-5.6 SolOpenAIReleased 2026-07-09 · Evidence-sensitiveCached input ¥2.68Input ¥26.82 · Output ¥134.0988.7
08Claude Fable 5AnthropicReleased 2026-06-09 · Evidence-sensitiveCached input ¥6.7Input ¥67.05 · Output ¥335.2388.4
09Gemini 3.6 FlashGoogleReleased 2026-07-21 · Evidence-sensitiveCached input ¥0.5Input ¥5.03 · Output ¥25.1486.1
10Gemini 3.5 FlashGoogleReleased 2026-05-19 · Evidence-sensitiveCached input ¥1.01Input ¥10.06 · Output ¥60.3485.0
11Gemini 3 Flash PreviewGoogleReleased 2025-12-17 · Evidence-sensitiveCached input ¥0.34Input ¥3.35 · Output ¥20.1184.2
12GPT-5.5OpenAIReleased 2026-04-23 · Evidence-sensitiveCached input ¥3.35Input ¥33.52 · Output ¥201.1480.2
13GPT-6 SolOpenAIReleased 2026-09-22 · Evidence-sensitiveCached input ¥1.34Input ¥13.41 · Output ¥67.0578.8
14Claude Opus 5AnthropicReleased 2026-07-24 · Evidence-sensitiveCached input ¥3.35Input ¥33.52 · Output ¥167.6176.7
15Qwen3.7 MaxAlibabaReleased 2026-05-19 · Evidence-sensitiveCached input ¥2.4Input ¥12 · Output ¥3668.7
16DeepSeek V4 Pro 0813DeepSeekReleased 2026-08-13 · Evidence-sensitiveCached input ¥0.3Input ¥9 · Output ¥2764.1
17Claude Opus 4.8AnthropicReleased 2026-05-28 · Evidence-sensitiveCached input ¥3.35Input ¥33.52 · Output ¥167.6163.7
18Kimi K3Moonshot AIReleased 2026-07-16 · Evidence-sensitiveCached input ¥2Input ¥20 · Output ¥10060.8
19Qwen3.6 Max PreviewAlibabaReleased 2026-04-20 · Evidence-sensitiveCached input ¥0.9Input ¥9 · Output ¥5460.7
20Claude Opus 4.7AnthropicReleased 2026-04-16 · Evidence-sensitiveCached input ¥3.35Input ¥33.52 · Output ¥167.6159.1
21Grok 4.6xAIReleased 2026-08-12 · Evidence-sensitiveCached input ¥3.35Input ¥13.41 · Output ¥40.2356.9
22GPT-5 2025-08-07OpenAIReleased 2025-08-07 · Evidence-sensitiveCached input ¥0.84Input ¥8.38 · Output ¥67.0556.3
23Grok 4.5xAIReleased 2026-07-08 · Evidence-sensitiveCached input ¥2.01Input ¥13.41 · Output ¥40.2355.7
24o3 2025-04-16OpenAIReleased 2025-04-16 · Evidence-sensitiveCached input ¥3.35Input ¥13.41 · Output ¥53.6454.8
25Qwen3.8 Max 0902AlibabaReleased 2026-09-01 · Evidence-sensitiveCached input —Input ¥12 · Output ¥3653.8
26GPT-5.1 2025-11-13OpenAIReleased 2025-11-13 · Evidence-sensitiveCached input ¥0.84Input ¥8.38 · Output ¥67.0553.3
27Qwen3 Max 2025-09-23AlibabaReleased 2025-09-24 · Evidence-sensitiveCached input —Input ¥6 · Output ¥2452.6
28claude-sonnet-5.5anthropicReleased — · Evidence-sensitiveCached input —Input — · Output —52.3
29GPT-5.4 Pro 2026-03-05OpenAIReleased 2026-03-05 · Evidence-sensitiveCached input —Input ¥201.14 · Output ¥1,206.8150.6
30DeepSeek V4 Pro PreviewDeepSeekReleased 2026-04-24 · Evidence-sensitiveCached input ¥0.0243Input ¥2.92 · Output ¥5.8348.9

Each board shows at most 30 models

The consensus index is not an accuracy rate; tied scores keep the order the shared evidence decides.

How to read this board

The current knowledge leaderboard uses two Epoch evaluations, both from the same institution. Prices play no part in the ranking, a missing evaluation does not count as zero, and the index is not an accuracy rate.

How it is calculated

About prices

API prices come from the providers' own sites, per million tokens. Dollar prices are converted to CNY at the rate of 2026-09-30. The cached price is the input price on a cache hit; cache writes, storage and subscriptions are extra.