CtrlAltDebrief

GPQA Diamond benchmark

449 AI models ranked on GPQA Diamond. Each row shows the model’s score next to its cheapest API price per 1M tokens where it has one, so you can weigh quality against cost.

Top model today

GPT-6 Astra (OpenAI) leads at 96%, from $10.00 per 1M input via OpenAI.

Data via arXiv:2311.12022 →

Models ranked

449

on this benchmark

Top score

96%

current leader

Type

Single eval

benchmark

Updated

Oct 14, 2026

benchmark scores

Cost vs GPQA Diamond

279 priced models
Most attractive quadrant (cheap + high)
= reasoning model (extended thinking)
0%20%40%60%80%100%$0.01$0.03$0.1$0.3$1$3$10$30COST / 1M INPUT TOKENS (log) →GPQA DIAMOND SCORE ↑GPT-6 AstraGemini 3.8 FlashGrok 4.6Gemini 3.7 FlashGemini 3.1 Pro PreviewGPT-5.6 SolClaude Fable 5.1Kimi K3MiniMax M3GLM 5.3DeepSeek V4 Flash 0731gpt-oss-20bQwen2.5 Coder 7B InstructLlama 3.2 3B Instruct

Each labelled name links to that model’s page. Curated to the cost-efficiency frontier, top scorers and cheapest so labels stay readable; the full field is in the leaderboard table below (and via “select all”). Reasoning models (extended “thinking” before answering) carry a ring around the dot.

Models ranked on GPQA Diamond

highest score first · 449 models
Maker
#
1GPT-6 AstraOpenAI96%
$10.00
via OpenAI
$50.00
via OpenAI
2Gemini 3.8 FlashGoogle95%
$0.75
via Google
$3.75
via Google
3Grok 4.6xAI95%
$2.00
via xAI
$6.00
via xAI
4Gemini 3.7 FlashGoogle95%
$0.75
via Google
$3.75
via Google
5Gemini 3.1 Pro PreviewGoogle94%
$2.00
via Google
$12.00
via Google
6GPT-5.6 SolOpenAI94%
$2.00
Routed quote
$10.00
Routed quote
7Claude Fable 5.1Anthropic94%
$10.00
via Anthropic
$50.00
via Anthropic
8Kimi K3Moonshot94%
$0.70
Routed quote
$12.50
via Sail
9GPT-5.5OpenAI94%
$5.00
via OpenAI
$30.00
via OpenAI
10Qwen3.8 2.4T A95BAlibaba94%
$2.00
via Aihubmix
$6.00
via Aihubmix
11Muse Spark 1.3Other94%
$1.25
via Azure
$4.25
via Azure
12Claude Opus 5Anthropic93%
$5.00
via Anthropic
$25.00
via Anthropic
13Grok 4.5xAI93%
$2.00
via xAI
$6.00
via xAI
14MiniMax M3MiniMax93%
$0.23
via Wandb
$0.96
via Wandb
15DeepSeek V4 Pro 0813DeepSeek93%
$0.66
Routed quote
$1.98
Routed quote
16Gemini 3.6 FlashGoogle93%
$0.75
via Google
$3.75
via Google
17Qwen3.8 Max (0902)Alibaba93%
$2.00
Routed quote
$6.00
Routed quote
18Qwen3.8 MaxAlibaba93%
$1.65
via DeepInfra
$4.95
via DeepInfra
19Claude Fable 5Anthropic93%
$10.00
via Anthropic
$50.00
via Anthropic
20GPT-5.6 TerraOpenAI93%
$2.00
via OpenAI
$12.00
via OpenAI
21Qwen3.7 MaxAlibaba92%
$1.25
via Novita
$3.75
via Novita
22Qwen3.8-Flash-NextAlibaba92%
n/a
n/a
23Gemini 3.5 FlashGoogle92%
$1.50
via Google
$9.00
via Google
24Claude Opus 4.8Anthropic92%
$5.00
via Anthropic
$25.00
via Anthropic
25GPT-5.4OpenAI92%
$2.50
via Databricks
$15.00
via OpenAI
26GLM 5.3Zhipu92%
$0.079
Routed quote
$3.08
via Sail
27GPT-5.3-CodexOpenAI92%
$1.75
via Aihubmix
$14.00
via Aihubmix
28Claude Opus 4.7Anthropic91%
$5.00
via Anthropic
$25.00
via Anthropic
29DeepSeek V4 Flash Vision (Reasoning, Max Effort)DeepSeek91%
n/a
n/a
30GLM 5.3 FlashZhipu91%
$0.11
via Sail
$0.35
via Sail
31Kimi K2.6Moonshot91%
$0.65
via Wandb
$3.40
via Novita
32Claude Sonnet 5Anthropic91%
$2.00
via Anthropic
$10.00
via Anthropic
33GPT-5.6 LunaOpenAI91%
$0.20
via OpenAI
$1.20
via OpenAI
34Grok 4.20xAI91%
$1.25
via xAI
$2.50
via xAI
35DeepSeek V4 Flash 0731DeepSeek91%
$0.016
Routed quote
$0.153
Routed quote
36Gemini 3 ProGoogle91%
$2.00
via Replicate
$12.00
via Replicate
37DeepSeek V4 Pro 0423DeepSeek91%high
$0.435
via Tencent
$0.87
via Tencent
38Qwen3.8 27BAlibaba91%
$0.04
Routed quote
$1.49
via Cerebras
39Agnes 2.5 Pro BetaSapiens-ai91%
n/a
n/a
40Muse Spark 1.2Other90%
$1.25
via Meta
$4.25
via Meta
41GPT-5.2OpenAI90%
$1.75
via OpenAI
$14.00
via OpenAI
42Grok 4.3xAI90%
$1.25
via xAI
$2.50
via xAI
43Qwen3.7 PlusAlibaba90%
$0.282
via Aihubmix
$1.13
via Aihubmix
44GPT-5.2-CodexOpenAI90%
$1.75
via Databricks
$14.00
via Databricks
45Muse Spark 1.1Other90%
$1.25
via Meta
$4.25
via Meta
46Gemini 3 FlashGoogle90%reasoning: true
$0.50
Routed quote
$3.00
Routed quote
47Hy3Tencent90%
$0.132
Routed quote
$0.528
Routed quote
48Kimi K2.7 CodeMoonshot90%
$0.671
Routed quote
$3.35
Routed quote
49Claude Opus 4.6Anthropic90%reasoning: adaptive
$5.00
via Anthropic
$25.00
via Anthropic
50GLM 5.2Zhipu89%
$0.05
Routed quote
$1.98
via Scx AI
1–50 of 461
Page 1 of 10
On-demand listed rates, OpenRouter routed quotes and Vercel reseller quotes. Enterprise commitments can be materially lower. Cost per task uses best-effort token usage. An OpenRouter routed quote is the price of whichever upstream provider OpenRouter prefers today, so it moves day to day.

Frequently asked questions

What is the GPQA Diamond benchmark?

A 198-question multiple-choice subset of GPQA covering graduate-level biology, physics and chemistry, written to be 'Google-proof' so answers can't be found by web search. Scored on accuracy: domain PhDs reach about 65%, while skilled non-experts manage only about 34% even with web access.

Which AI model scores highest on GPQA Diamond?

GPT-6 Astra (OpenAI) leads with 96%, from $10.00 per 1M input tokens from OpenAI.

How many models are ranked on GPQA Diamond?

449 models carry a GPQA Diamond score. Scores come from Artificial Analysis; where a model has a listed API price it is drawn from LiteLLM and OpenRouter, and models with no current price are listed without one.

Other benchmarks

compare the same models on a different eval
Intelligence Indexcomposite indexCoding Indexcomposite indexAgentic Indexcomposite indexHumanity's Last ExamleaderboardMMLU-ProleaderboardMMMU-ProleaderboardLiveCodeBenchleaderboardSciCodeleaderboardTerminalBench HardleaderboardTerminalBench v2.1leaderboardτ²-benchleaderboardτ-bench BankingleaderboardAPEX AgentsleaderboardIT-Bench SREleaderboardGDPvalleaderboardCritPtleaderboardOmniscienceleaderboardIFBenchleaderboardLong-Context Reasoningcomposite indexSWE-bench VerifiedleaderboardSWE-bench ProleaderboardAIME 2026leaderboardARC-AGI-2leaderboardAA-Briefcaseleaderboard
Benchmark scores via Artificial Analysis · scores updated Oct 14, 2026 · Listed prices from LiteLLM, OpenRouter and Vercel · catalogue refreshed Oct 14, 2026 · 160 prices older, back to Jul 19, 2026

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI