AI Model Leaderboard

Benchmark scores across leading AI models. Click column headers to sort.

Benchmark:44 results

Rank ↕	Model ↕	Company ↕	Benchmark ↕	Score ↓
1	Claude Opus 4.7	Anthropic	Arena ELO	1285.0
2	Claude Opus 4.6	Anthropic	Arena ELO	1270.0
4	GPT-4o	OpenAI	Arena ELO	1250.0
5	Gemini 3.1 Pro	Google DeepMind	Arena ELO	1235.0
3	Claude Sonnet 4.6	Anthropic	Arena ELO	1220.0
6	Gemini 2.5 Flash	Google DeepMind	Arena ELO	1180.0
1	Claude Opus 4.7	Anthropic	HumanEval	97.2
2	Claude Opus 4.6	Anthropic	HumanEval	96.5
10	GPT-5.4 Thinking	OpenAI	HumanEval	95.0
3	Claude Sonnet 4.6	Anthropic	HumanEval	94.0
4	GPT-4o	OpenAI	HumanEval	93.5
1	Claude Opus 4.7	Anthropic	GPQA	93.1
2	Claude Opus 4.6	Anthropic	GPQA	92.5
1	Claude Opus 4.7	Anthropic	MMLU	92.3
5	Gemini 3.1 Pro	Google DeepMind	HumanEval	92.1
10	GPT-5.4 Thinking	OpenAI	GPQA	92.0
4	GPT-4o	OpenAI	GPQA	91.8
1	Claude Opus 4.7	Anthropic	MATH	91.5
2	Claude Opus 4.6	Anthropic	MMLU	91.2
6	Gemini 2.5 Flash	Google DeepMind	HumanEval	91.0
5	Gemini 3.1 Pro	Google DeepMind	GPQA	90.5
3	Claude Sonnet 4.6	Anthropic	GPQA	90.2
2	Claude Opus 4.6	Anthropic	MATH	90.1
7	Llama 3.2 405B	Meta AI	HumanEval	90.0
10	GPT-5.4 Thinking	OpenAI	MMLU	90.0
6	Gemini 2.5 Flash	Google DeepMind	GPQA	89.0
10	GPT-5.4 Thinking	OpenAI	MATH	89.0
7	Llama 3.2 405B	Meta AI	GPQA	88.8
4	GPT-4o	OpenAI	MMLU	88.7
4	GPT-4o	OpenAI	MATH	88.2
3	Claude Sonnet 4.6	Anthropic	MMLU	88.1
8	Grok 3	xAI	HumanEval	88.0
3	Claude Sonnet 4.6	Anthropic	MATH	87.5
9	DeepSeek R1	—	HumanEval	87.0
5	Gemini 3.1 Pro	Google DeepMind	MMLU	86.5
7	Llama 3.2 405B	Meta AI	MMLU	85.2
5	Gemini 3.1 Pro	Google DeepMind	MATH	85.0
7	Llama 3.2 405B	Meta AI	MATH	83.5
6	Gemini 2.5 Flash	Google DeepMind	MMLU	83.0
6	Gemini 2.5 Flash	Google DeepMind	MATH	82.0
8	Grok 3	xAI	MMLU	80.1
8	Grok 3	xAI	MATH	79.5
9	DeepSeek R1	—	MMLU	78.5
9	DeepSeek R1	—	MATH	77.0