SimpleBench

Measures: 200+ multiple-choice trick questions on spatio-temporal reasoning, social intelligence and linguistic adversarial robustness that non-experts answer easily; scored as average accuracy over 5 runs (AVG@5).

Maintainer: SimpleBench team (AI Explained)
Introduced: 2024
Status: Active
Saturation: Near-ceiling
Typical provenance: Independent
In scoring set: Yes

Official leaderboard

Leaderboard

Show all results

Independent and third-party results only.

Leaderboard
Model	Score (AVG@5)	Provenance	Source	Evaluated
GPT-4o mini OpenAI	10.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
Command R+ (08-2024) Cohere	17.4%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-4o OpenAI	17.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
o1-mini OpenAI · Medium effort	18.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
DeepSeek-V3 DeepSeek	18.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.0 Flash Google DeepMind	18.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
Llama 3.3 70B Meta	19.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
gpt-oss-120b OpenAI	22.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
Mistral Large 2.0 Mistral AI	22.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
Grok 2 SpaceXAI (xAI)	22.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
o3-mini OpenAI · High effort	22.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
Llama 3.1 405B Meta	23.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude 3 Opus Anthropic	23.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-4 Turbo OpenAI	25.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-4.1 OpenAI	27.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 1.5 Pro Google DeepMind	27.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude 3.5 Sonnet Anthropic	27.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
Llama 4 Maverick Meta	27.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.0 Flash Thinking Google DeepMind	30.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
DeepSeek-R1 DeepSeek	30.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
Qwen3-235B-A22B Alibaba Group	31.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.0 Pro Google DeepMind	31.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-4.5 OpenAI	34.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
Qwen3.6-Flash Alibaba Group	35.2%	Third party	SimpleBench Leaderboard via Epoch AI	—
Grok 3 SpaceXAI (xAI)	36.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
o1 OpenAI · Medium effort	36.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
o4-mini OpenAI · High effort	38.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
DeepSeek-V3.1 DeepSeek	40.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
o1 OpenAI · High effort	40.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
DeepSeek-R1 DeepSeek	40.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.5 Flash Google DeepMind	41.2%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude 3.5 Sonnet Anthropic	41.4%	Third party	SimpleBench Leaderboard via Epoch AI	—
o1-preview OpenAI	41.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude 3.7 Sonnet Anthropic	44.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Sonnet 4 Anthropic · 12K effort	45.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Sonnet 4 Anthropic · Unknown effort	45.5%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.2 OpenAI · High effort	45.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.2 OpenAI · Unknown effort	45.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
MiniMax-M3 MiniMax	45.8%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
DeepSeek-V4-Flash DeepSeek · Unknown effort	46.3%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude 3.7 Sonnet Anthropic · 12K effort	46.4%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude 3.7 Sonnet Anthropic · Unknown effort	46.4%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Luna OpenAI · Unknown effort	46.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Luna OpenAI · Xhigh effort	46.8%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Kimi K2.5 Moonshot AI	46.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
GLM-4.7 Z.ai (Zhipu AI)	47.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
GLM-4.7 Z.ai (Zhipu AI)	47.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Terra OpenAI · Unknown effort	48.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Terra OpenAI · Xhigh effort	48.9%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Inkling Thinking Machines Lab	50.0%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Gemini 2.5 Pro Google DeepMind	51.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.5 Pro Google DeepMind	51.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
DeepSeek-V3.2-Speciale DeepSeek	52.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
o3 OpenAI · High effort	53.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
GLM-5 Z.ai (Zhipu AI)	53.2%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.1 OpenAI · High effort	53.2%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Sonnet 4.5 Anthropic · 12K effort	54.3%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Sonnet 4.5 Anthropic · Unknown effort	54.3%	Third party	SimpleBench Leaderboard via Epoch AI	—
GLM-5.1 Z.ai (Zhipu AI)	55.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
Grok 4.1 Fast SpaceXAI (xAI)	56.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Grok 4.1 Fast SpaceXAI (xAI)	56.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5 OpenAI · High effort	56.7%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5.2 Pro OpenAI	57.4%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Kimi K2.7 Code Moonshot AI	57.9%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Opus 4 Anthropic · 12K effort	58.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Opus 4 Anthropic · Unknown effort	58.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
GLM-5.2 Z.ai (Zhipu AI) · Unknown effort	58.8%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Opus 4.1 Anthropic	60.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Opus 4.1 Anthropic · Unknown effort	60.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Grok 4 SpaceXAI (xAI)	60.5%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Sonnet 5 Anthropic · Unknown effort	60.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
Kimi K3 Moonshot AI · Max effort	60.7%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
DeepSeek-V4-Flash DeepSeek · Unknown effort	61.1%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Gemini 3 Flash Google DeepMind	61.1%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5 Pro OpenAI · High effort	61.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5 Pro OpenAI · Unknown effort	61.6%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Opus 4.7 Anthropic · Unknown effort	61.7%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Opus 4.5 Anthropic	62.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Opus 4.5 Anthropic · Unknown effort	62.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 2.5 Pro Google DeepMind	62.4%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Qwen3.6-Max-Preview Alibaba Group	63.0%	Third party	SimpleBench Leaderboard via Epoch AI	—
Claude Opus 4.8 Anthropic · Unknown effort	64.8%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5.6 Sol OpenAI · Unknown effort	64.8%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Sol OpenAI · Xhigh effort	64.8%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Opus 4.6 Anthropic	67.6%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5.5 OpenAI · Unknown effort	69.0%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Grok 4.5 SpaceXAI (xAI) · Unknown effort	70.0%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Qwen3.7-Max Alibaba Group	70.4%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.6 Sol OpenAI · Proxhigh effort	71.7%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5.6 Sol OpenAI · Prounknown effort	71.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.4 Pro OpenAI · Unknown effort	74.1%	Third party	SimpleBench Leaderboard via Epoch AI	—
Muse Spark 1.2 Meta	74.5%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Grok 4.6 SpaceXAI (xAI) · Unknown effort	75.9%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Gemini 3 Pro Google DeepMind	76.4%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Gemini 3.5 Flash Google DeepMind · Unknown effort	76.7%	Third party	SimpleBench Leaderboard via Epoch AI	—
GPT-5.5 Pro OpenAI · Unknown effort	76.9%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
GPT-5.5 Pro OpenAI · High effort	76.9%	Third party	SimpleBench Leaderboard via Epoch AI	—
Gemini 3.1 Pro Google DeepMind	79.6%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Opus 5 Anthropic · Unknown effort	80.6%	Third party	https://lmcouncil.ai/benchmarks via Epoch AI	—
Claude Fable 5 Anthropic · Max effort	81.9%	Third party	SimpleBench Leaderboard via Epoch AI	—

Epoch AI, 'Capabilities & benchmarking', epoch.ai/benchmarks, CC BY 4.0.

Sources

Epoch AI