Scientific Reasoning Leaderboard

Models ranked by GPQA Diamond, based on independent benchmark evaluations.

Each row shows the model's benchmark score alongside its pricing and output speed, so you can evaluate quality-to-cost tradeoffs at a glance.

Top 20 models ranked by gpqa diamond

Rank	Model	Creator	GPQA Diamond	Price/1M	Speed
🥇	Google: Gemini 3.1 Pro Preview Google	Google	0.9	$4.50	120 tok/s
🥈	OpenAI: GPT-5.5 OpenAI	OpenAI	0.9	$11.25	70 tok/s
🥉	GPT-5.5 (high) OpenAI	OpenAI	0.9	$11.25	67 tok/s
4	GPT-5.5 (medium) OpenAI	OpenAI	0.9	$11.25	64 tok/s
5	OpenAI: GPT-5.4 OpenAI	OpenAI	0.9	$5.63	76 tok/s
6	OpenAI: GPT-5.3-Codex OpenAI	OpenAI	0.9	$4.81	80 tok/s
7	Anthropic: Claude Opus 4.7 Anthropic	Anthropic	0.9	$10.00	55 tok/s
8	MoonshotAI: Kimi K2.6 Kimi	Kimi	0.9	$1.44	38 tok/s
9	xAI: Grok 4.20 xAI	xAI	0.9	$1.56	98 tok/s
10	GPT-5.5 (low) OpenAI	OpenAI	0.9	$11.25	64 tok/s
11	Gemini 3 Pro Preview (high) Google	Google	0.9	$4.50	109 tok/s
12	DeepSeek V4 Pro (Reasoning, High Effort) DeepSeek	DeepSeek	0.9	$2.17	33 tok/s
13	OpenAI: GPT-5.2 OpenAI	OpenAI	0.9	$4.81	70 tok/s
14	xAI: Grok 4.3 xAI	xAI	0.9	$1.56	82 tok/s
15	OpenAI: GPT-5.2-Codex OpenAI	OpenAI	0.9	$4.81	96 tok/s
16	Gemini 3 Flash Preview (Reasoning) Google	Google	0.9	$1.13	198 tok/s
17	Claude Opus 4.6 (Adaptive Reasoning, Max Effort) Anthropic	Anthropic	0.9	$10.94	50 tok/s
18	DeepSeek: DeepSeek V4 Flash DeepSeek	DeepSeek	0.9	$0.18	72 tok/s
19	Qwen: Qwen3.5 397B A17B Alibaba	Alibaba	0.9	$0.88	52 tok/s
20	Qwen3.6 Max Preview Alibaba	Alibaba	0.9	$2.92	37 tok/s