Skip to content
Weights Not Walls
Menu

SimpleBench

Reported

No description yet: we only describe an eval set when we can do it correctly.

Kind
Published by a third party's leaderboard, collected by Epoch AI. The leaderboard is named and linked.
Link
https://simple-bench.com/
Score shown
Epoch's column "Score (AVG@5)", as a percentage, higher is better; random guessing scores about 16.7%. The source gives no error bars.
Results
125 in all: 34 for open-weight models, 89 for closed ones, 2 where Epoch does not say. 31 linked to a model page here.

Best open-weight result: deepseek-ai/DeepSeek-V4.1-Flash, 66.7%, number 31 of 125 results.

SimpleBench results
#ModelWeightsSettingScoreDateSource
1claude-fable-5-2_unknownAnthropicclosedeffort unknown94.9%not datedsimple-bench.com
2gemini-4-carbon_unknownGoogle DeepMindclosedeffort unknown92.6%not datedsimple-bench.com
3gemini-4-argon_unknownGoogle DeepMindclosedeffort unknown91.7%not datedsimple-bench.com
4claude-opus-5-5_unknownAnthropicclosedeffort unknown88.4%not datedsimple-bench.com
5claude-fable-5-1_unknownAnthropicclosedeffort unknown86.6%not datedsimple-bench.com
6gpt-6-astra_promaxOpenAIcloseddefault86.5%not datedsimple-bench.com
7gpt-6-astra_unknownOpenAIclosedeffort unknown83.6%not datedsimple-bench.com
8gpt-6.1-sol_unknownOpenAIclosedeffort unknown82.9%not datedsimple-bench.com
9gemini-3.8-flash_unknownGoogle DeepMindclosedeffort unknown82.4%not datedsimple-bench.com
10claude-fable-5_maxAnthropicclosedeffort max81.9%not datedsimple-bench.com
11muse-spark-1.3_unknownMeta AIclosedeffort unknown81.8%not datedsimple-bench.com
12claude-opus-5_unknownAnthropicclosedeffort unknown80.6%not datedlmcouncil.ai
13gemini-3.1-pro-previewGoogle DeepMindcloseddefault79.6%not datedlmcouncil.ai
14gpt-5.5-pro_highOpenAIclosedeffort high76.9%not datedSimpleBench Leaderboard, via Epoch AI
15gpt-5.5-pro_unknownOpenAIclosedeffort unknown76.9%not datedlmcouncil.ai
16gemini-3.5-flash_unknownGoogle DeepMindclosedeffort unknown76.7%not datedSimpleBench Leaderboard, via Epoch AI
17gemini-3-pro-previewGoogle DeepMindcloseddefault76.4%not datedlmcouncil.ai
18claude-sonnet-5-5_unknownAnthropicclosedeffort unknown75.9%not datedsimple-bench.com
19grok-4.6_unknownxAIclosedeffort unknown75.9%not datedlmcouncil.ai
20muse-spark-1.2Meta AIcloseddefault74.5%not datedlmcouncil.ai
21gpt-5.4-pro-2026-03-05_unknownOpenAIclosedeffort unknown74.1%not datedSimpleBench Leaderboard
22gpt-6-sol_unknownOpenAIclosedeffort unknown73.1%not datedsimple-bench.com
23muse-ember-1.0_unknownMeta AIclosedeffort unknown72.7%not datedsimple-bench.com
24gpt-6.1-terra_unknownOpenAIclosedeffort unknown72.2%not datedsimple-bench.com
25gpt-5.6-sol_prounknownOpenAIcloseddefault71.7%not datedSimpleBench Leaderboard, via Epoch AI
26gpt-5.6-sol_proxhighOpenAIcloseddefault71.7%not datedlmcouncil.ai
27qwen3.7-maxAlibabacloseddefault70.4%not datedSimpleBench Leaderboard, via Epoch AI
28grok-4.5_unknownxAIclosedeffort unknown70.0%not datedlmcouncil.ai
29gpt-5.5_unknownOpenAIclosedeffort unknown69.0%not datedlmcouncil.ai
30claude-opus-4-6Anthropiccloseddefault67.6%not datedlmcouncil.ai
31deepseek-ai/DeepSeek-V4.1-FlashEpoch: deepseek-v4.1-flash_unknown, DeepSeekopeneffort unknown66.7%not datedsimple-bench.com
32zai-org/GLM-5.3Epoch: glm-5.3_unknown, Z.ai (Zhipu AI)openeffort unknown66.2%not datedsimple-bench.com
33claude-opus-4-8_unknownAnthropicclosedeffort unknown64.8%not datedlmcouncil.ai
34gpt-5.6-sol_unknownOpenAIclosedeffort unknown64.8%not datedSimpleBench Leaderboard, via Epoch AI
35gpt-5.6-sol_xhighOpenAIclosedeffort xhigh64.8%not datedlmcouncil.ai
36qwen3.6-max-previewAlibabanot stateddefault63.0%not datedSimpleBench Leaderboard, via Epoch AI
37gemini-2.5-pro-preview-06-05Google DeepMindcloseddefault62.4%not datedlmcouncil.ai
38claude-opus-4-5-20251101Anthropiccloseddefault62.0%not datedSimpleBench Leaderboard
39claude-opus-4-5-20251101_unknownAnthropicclosedeffort unknown62.0%not datedSimpleBench Leaderboard, via Epoch AI
40claude-opus-4-7_unknownAnthropicclosedeffort unknown61.7%not datedlmcouncil.ai
41gpt-5-pro-2025-10-06_highOpenAIclosedeffort high61.6%not datedSimpleBench Leaderboard
42gpt-5-pro-2025-10-06_unknownOpenAIclosedeffort unknown61.6%not datedSimpleBench Leaderboard, via Epoch AI
43deepseek-ai/DeepSeek-V4-Flash-0731Epoch: deepseek-v4-flash-0731_unknown, DeepSeekopeneffort unknown61.1%not datedlmcouncil.ai
44gemini-3-flash-previewGoogle DeepMindcloseddefault61.1%not datedlmcouncil.ai
45moonshotai/Kimi-K3Epoch: kimi-k3_max, Moonshotopeneffort max60.7%not datedlmcouncil.ai
46claude-sonnet-5_unknownAnthropicclosedeffort unknown60.6%not datedSimpleBench Leaderboard, via Epoch AI
47grok-4-0709xAIcloseddefault60.5%not datedlmcouncil.ai
48Qwen/Qwen3.8-27BEpoch: qwen3.8-27b, Alibabaopendefault60.2%not datedsimple-bench.com
49claude-opus-4-1-20250805Anthropiccloseddefault60.0%not datedSimpleBench Leaderboard
50claude-opus-4-1-20250805_unknownAnthropicclosedeffort unknown60.0%not datedSimpleBench Leaderboard, via Epoch AI
51claude-haiku-5-5_unknownAnthropicclosedeffort unknown59.2%not datedsimple-bench.com
52claude-opus-4-20250514_12KAnthropiccloseddefault58.8%not datedSimpleBench Leaderboard
53claude-opus-4-20250514_unknownAnthropicclosedeffort unknown58.8%not datedSimpleBench Leaderboard, via Epoch AI
54zai-org/GLM-5.2Epoch: glm-5.2_unknown, Z.ai (Zhipu AI)openeffort unknown58.8%not datedlmcouncil.ai
55moonshotai/Kimi-K2.7-CodeEpoch: kimi-k2.7-code, Moonshotopendefault57.9%not datedlmcouncil.ai
56gpt-5.2-pro-2025-12-11OpenAIcloseddefault57.4%not datedlmcouncil.ai
57gpt-5-2025-08-07_highOpenAIclosedeffort high56.7%not datedlmcouncil.ai
58grok-4-1-fast-non-reasoningxAIcloseddefault56.0%not datedSimpleBench Leaderboard
59grok-4-1-fast-reasoningxAIcloseddefault56.0%not datedSimpleBench Leaderboard, via Epoch AI
60zai-org/GLM-5.1Epoch: glm-5.1, Z.ai (Zhipu AI)opendefault55.1%not datedSimpleBench Leaderboard, via Epoch AI
61claude-sonnet-4-5-20250929_12KAnthropiccloseddefault54.3%not datedSimpleBench Leaderboard
62claude-sonnet-4-5-20250929_unknownAnthropicclosedeffort unknown54.3%not datedSimpleBench Leaderboard, via Epoch AI
63zai-org/GLM-5Epoch: glm-5, Z.ai (Zhipu AI)opendefault53.2%not datedSimpleBench Leaderboard, via Epoch AI
64gpt-5.1-2025-11-13_highOpenAIclosedeffort high53.2%not datedlmcouncil.ai
65o3-2025-04-16_highOpenAIclosedeffort high53.1%not datedSimpleBench Leaderboard
66deepseek-ai/DeepSeek-V3.2-SpecialeEpoch: DeepSeek-V3.2-Speciale, DeepSeekopendefault52.6%not datedSimpleBench Leaderboard, via Epoch AI
67gemini-2.5-pro-exp-03-25Google DeepMindcloseddefault51.6%not datedSimpleBench Leaderboard
68gemini-2.5-pro-preview-03-25Google DeepMindcloseddefault51.6%not datedSimpleBench Leaderboard, via Epoch AI
69thinkingmachines/InklingEpoch: Inkling, Thinking Machinesopendefault50.0%not datedlmcouncil.ai
70gpt-5.6-terra_unknownOpenAIclosedeffort unknown48.9%not datedSimpleBench Leaderboard, via Epoch AI
71gpt-5.6-terra_xhighOpenAIclosedeffort xhigh48.9%not datedlmcouncil.ai
72zai-org/GLM-4.7Epoch: glm-4.7, Z.ai (Zhipu AI)opendefault47.7%not datedSimpleBench Leaderboard
73zai-org/GLM-4.7Epoch: zai-org/glm-4.7, Z.ai (Zhipu AI)opendefault47.7%not datedSimpleBench Leaderboard, via Epoch AI
74gpt-5.6-luna_unknownOpenAIclosedeffort unknown46.8%not datedSimpleBench Leaderboard, via Epoch AI
75gpt-5.6-luna_xhighOpenAIclosedeffort xhigh46.8%not datedlmcouncil.ai
76moonshotai/Kimi-K2.5Epoch: kimi-k2.5, Moonshotopendefault46.8%not datedSimpleBench Leaderboard, via Epoch AI
77claude-3-7-sonnet-20250219_12KAnthropiccloseddefault46.4%not datedSimpleBench Leaderboard
78claude-3-7-sonnet-20250219_unknownAnthropicclosedeffort unknown46.4%not datedSimpleBench Leaderboard, via Epoch AI
79deepseek-ai/DeepSeek-V4-FlashEpoch: deepseek-v4-flash_unknown, DeepSeekopeneffort unknown46.3%not datedlmcouncil.ai
80gpt-5.2-2025-12-11_highOpenAIclosedeffort high45.8%not datedSimpleBench Leaderboard
81gpt-5.2-2025-12-11_unknownOpenAIclosedeffort unknown45.8%not datedSimpleBench Leaderboard, via Epoch AI
82MiniMaxAI/MiniMax-M3Epoch: MiniMax-M3, MiniMaxopendefault45.8%not datedlmcouncil.ai
83claude-sonnet-4-20250514_12KAnthropiccloseddefault45.5%not datedSimpleBench Leaderboard
84claude-sonnet-4-20250514_unknownAnthropicclosedeffort unknown45.5%not datedSimpleBench Leaderboard, via Epoch AI
85claude-3-7-sonnet-20250219Anthropiccloseddefault44.9%not datedSimpleBench Leaderboard
86nemotron-3-ultraNvidiano page matchopendefault41.7%not datedsimple-bench.com
87o1-preview-2024-09-12OpenAIcloseddefault41.7%not datedSimpleBench Leaderboard
88claude-3-5-sonnet-20241022Anthropiccloseddefault41.4%not datedSimpleBench Leaderboard
89gemini-2.5-flashGoogle DeepMindcloseddefault41.2%not datedsimple-bench.com
90deepseek-ai/DeepSeek-R1-0528Epoch: DeepSeek-R1-0528, DeepSeekopendefault40.8%not datedsimple-bench.com
91o1-2024-12-17_highOpenAIclosedeffort high40.1%not datedSimpleBench Leaderboard
92deepseek-ai/DeepSeek-V3.1Epoch: DeepSeek-V3.1, DeepSeekopendefault40.0%not datedsimple-bench.com
93moonshotai/Kimi-K2-ThinkingEpoch: kimi-k2-thinking, Moonshotopendefault39.6%not datedsimple-bench.com
94o4-mini-2025-04-16_highOpenAIclosedeffort high38.7%not datedSimpleBench Leaderboard
95o1-2024-12-17_mediumOpenAIclosedeffort medium36.7%not datedSimpleBench Leaderboard
96grok-3xAIcloseddefault36.1%not datedSimpleBench Leaderboard
97qwen3.6-flashAlibabanot stateddefault35.2%not datedsimple-bench.com
98MiniMaxAI/MiniMax-M2.1Epoch: MiniMax-M2.1, MiniMaxopendefault34.7%not datedsimple-bench.com
99gpt-4.5-preview-2025-02-27OpenAIcloseddefault34.5%not datedSimpleBench Leaderboard
100gemini-exp-1206Google DeepMind,Googlecloseddefault31.1%not datedSimpleBench Leaderboard
101Qwen/Qwen3-235B-A22BEpoch: qwen3-235b-a22b, Alibabaopendefault31.0%not datedSimpleBench Leaderboard
102deepseek-ai/DeepSeek-R1Epoch: DeepSeek-R1, DeepSeekopendefault30.9%not datedSimpleBench Leaderboard
103gemini-2.0-flash-thinking-exp-01-21Google DeepMind,Googlecloseddefault30.7%not datedSimpleBench Leaderboard
104meta-llama/Llama-4-Maverick-17B-128E-InstructEpoch: Llama-4-Maverick-17B-128E-Instruct, Meta AIopendefault27.7%not datedSimpleBench Leaderboard
105claude-3-5-sonnet-20240620Anthropiccloseddefault27.5%not datedSimpleBench Leaderboard
106deepseek-ai/DeepSeek-V3-0324Epoch: DeepSeek-V3-0324, DeepSeekopendefault27.2%not datedSimpleBench Leaderboard
107gemini-1.5-pro-002Google DeepMindcloseddefault27.1%not datedSimpleBench Leaderboard
108gpt-4.1-2025-04-14OpenAIcloseddefault27.0%not datedSimpleBench Leaderboard
109moonshotai/Kimi-K2-InstructEpoch: Kimi-K2-Instruct, Moonshotopendefault26.3%not datedsimple-bench.com
110gpt-4-turbo-2024-04-09OpenAIcloseddefault25.1%not datedSimpleBench Leaderboard
111MiniMaxAI/MiniMax-M2Epoch: MiniMax-M2, MiniMaxopendefault25.0%not datedsimple-bench.com
112claude-3-opus-20240229Anthropiccloseddefault23.5%not datedSimpleBench Leaderboard
113meta-llama/Llama-3.1-405B-InstructEpoch: Llama-3.1-405B-Instruct, Meta AIopendefault23.0%not datedSimpleBench Leaderboard
114o3-mini-2025-01-31_highOpenAIclosedeffort high22.8%not datedSimpleBench Leaderboard
115grok-2-1212xAIcloseddefault22.7%not datedSimpleBench Leaderboard
116mistral-large-2407Mistral AIno page matchopendefault22.5%not datedSimpleBench Leaderboard
117openai/gpt-oss-120bEpoch: gpt-oss-120b, OpenAIopendefault22.1%not datedsimple-bench.com
118mistral-large-2512Mistral AIno page matchopendefault20.4%not datedsimple-bench.com
119meta-llama/Llama-3.3-70B-InstructEpoch: Llama-3.3-70B-Instruct, Meta AIopendefault19.9%not datedSimpleBench Leaderboard
120deepseek-ai/DeepSeek-V3Epoch: DeepSeek-V3, DeepSeekopendefault18.9%not datedSimpleBench Leaderboard
121gemini-2.0-flash-expGoogle DeepMind,Googlecloseddefault18.9%not datedSimpleBench Leaderboard
122o1-mini-2024-09-12_mediumOpenAIclosedeffort medium18.1%not datedSimpleBench Leaderboard
123gpt-4o-2024-08-06OpenAIcloseddefault17.8%not datedSimpleBench Leaderboard
124CohereLabs/c4ai-command-r-plus-08-2024Epoch: c4ai-command-r-plus-08-2024, Cohere,Cohere Labs (formerly Cohere for AI)opendefault17.4%not datedSimpleBench Leaderboard
125gpt-4o-mini-2024-07-18OpenAIcloseddefault10.7%not datedSimpleBench Leaderboard

Open-weight rows are shaded. Dates are when the result was run or evaluated, where the data says. Data from Epoch AI (CC-BY 4.0), fetched 19 Nov 2026: Epoch AI, ‘Capabilities & benchmarking’. Published online at epoch.ai. Retrieved from ‘https://epoch.ai/benchmarks’ [online resource]. Epoch AI. Method