Note: Overall leaderboard rankings may not reflect true model quality — individual benchmarks give a clearer picture. ARC-Challenge MMLU GPQA GSM8K Artificial Analysis Intelligence Index v4.1
← Back to leaderboard

OpenAI: o1

8 benchmarks
CompanyOpenAI
Country of originUSA

Leaderboard position

#402
Overall rank (of 1121 models)
1309
Elo rating
8
Benchmarks
Top 35.9%
Percentile
#36
In OpenAI (of 74 models)
Above: inclusionAI: ling-2.6-flash #401 Below: Xiaomi: mimo-v2-flash #403

Benchmark highlights

Best benchmark
GSM8K 97.1
Worst benchmark
POLLUX 1.2

Nearest competitor from another company

The model from another company with the most similar benchmark scores.

appeared as nearest in 2 of 8 benchmarks

Google: Gemma 4 26B A4B Instruct
2/8 25.0%
Model Votes % of benchmarks
Google: Gemma 4 26B A4B Instruct 2 25.0%
Xiaomi: mimo-v2-flash 1 12.5%
Z.ai: glm-4.7-flash 1 12.5%
Mistral AI: ministral-3-14b-instruct-2512 1 12.5%
Meta Llama: Llama 3.1 405B Instruct 1 12.5%
Google: Gemma 3 12B Instruct 1 12.5%
Anthropic: claude-3.5-sonnet 1 12.5%

Benchmarks

GSM8K 97.1 MATH 96.4 ChemBench 64 Artificial Analysis Agentic Index (Maximum Reasoning) 31.1 Artificial Analysis Intelligence Index v4.0 (Maximum Reasoning) 30.7 Artificial Analysis Intelligence Index v4.1 (Maximum Reasoning) 23 Artificial Analysis Coding Index (Maximum Reasoning) 20.5 POLLUX 1.2

Predicted scores

Estimated from 25 similar models and the global Elo rating. These are projections, not measured results.

Benchmark Predicted Confidence Based on
AIME 2026 ~82.29 60% Z.ai: GLM 5.1, Moonshot AI: Kimi K2.6
HMMT Feb 2026 ~73.83 60% Qwen: qwen3-30b-a3b, Z.ai: GLM 5.1, Moonshot AI: Kimi K2.6
GPQA Diamond ~67.68 60% OpenAI: o3-mini, OpenAI: gpt-oss-20b, Mistral AI: ministral-3-14b-instruct-2512
MMLU-Pro ~71.82 60% Google: Gemma 4 26B A4B Instruct, DeepSeek: deepseek-v3, IBM Granite: granite-4.1-30b
Humanity's Last Exam (no tools) ~18.99 60% OpenAI: o3-mini, OpenAI: gpt-oss-20b, Qwen: Qwen3 8B
IFEval ~88.49 60% OpenAI: o3-mini, Qwen: Qwen3 8B, Google: Gemma 3 27B Instruct
MBPP ~93.7 60% OpenAI: gpt-4o, Qwen: qwen3-1.7b
ARC-Easy ~94.8 60% Meta Llama: Llama 3.2 3B Instruct, Google: Gemma 3 270M Instruct
PIQA ~84.39 60% Google: Gemma 3 270M Instruct
SuperGPQA ~50.61 60% OpenAI: o3-mini, OpenAI: gpt-oss-20b, Meta Llama: Llama 3.1 405B Instruct
TruthfulQA ~54.43 60% Meta Llama: Llama 3.2 3B Instruct, Google: Gemma 3 270M Instruct
AIME 2025 ~80.46 59% OpenAI: gpt-oss-20b, Mistral AI: mistral-medium-3.5-128b, Mistral AI: ministral-3-14b-instruct-2512
MMLU ~77.99 59% OpenAI: gpt-4o, Meta Llama: Llama 3.2 3B Instruct, DeepSeek: DeepSeek R1
SWE-bench Verified ~53.92 59% OpenAI: gpt-oss-20b, Mistral AI: mistral-medium-3.5-128b, Qwen: Qwen3 8B
SWE-bench Pro ~31.55 59% Meta Llama: Llama 3.1 405B Instruct, Meta Llama: Llama 4 Maverick Instruct, Google: Gemma 3 27B Instruct
SciCode ~34.53 59% OpenAI: o3-mini, Technology Innovation Institute: falcon-h1r-7b, Meta Llama: Llama 3.3 70B Instruct
Humanity's Last Exam (with tools) ~27.33 59% OpenAI: gpt-oss-20b, Z.ai: GLM 5.1, Meta: Muse Spark
Text Arena (Snapshot from 20 june 2026) ~1361.57 59% OpenAI: gpt-oss-20b, Meta Llama: Llama 3.1 405B Instruct, Meta Llama: Llama 4 Maverick Instruct
MATH-500 ~93.76 58% Qwen: Qwen3 8B, Technology Innovation Institute: falcon-h1r-7b, Google: Gemma 3 4B Instruct
IFBench ~60.07 58% Mistral AI: mistral-medium-3.5-128b, Technology Innovation Institute: falcon-h1r-7b, Meta Llama: Llama 3.3 70B Instruct
HumanEval ~86.81 58% OpenAI: gpt-4o, DeepSeek: deepseek-v2.5, OpenAI: GPT-3.5 Turbo
MMMU-Pro ~62.5 58% SpaceXAI: Grok 4.3, Meta: Muse Spark, Moonshot AI: Kimi K2.6
Artificial Analysis Intelligence Index v4.3.2 ~20.37 58% SpaceXAI: Grok 4.1 Fast, Qwen: Qwen3.7 Max, Google: Gemma 3 270M Instruct
BrowseComp ~52.77 58% DeepSeek: DeepSeek V4 Pro, Z.ai: GLM 5.1, Moonshot AI: Kimi K2.6
GPQA ~61.23 58% OpenAI: gpt-4o, Qwen: Qwen3 8B, Meta Llama: Llama 3.2 3B Instruct