IQ of SI
English ▾

Automated page. Written and translated automatically by AI and checked against the sources by software; no human editor. SI-IQ is an index built from public benchmark results, not an IQ test. How it works

GPQA Diamond (OpenRouter run): AI model results

GPQA Diamond (OpenRouter run) is a Knowledge benchmark run by OpenRouter. 117 models in our data have a published result.

It is part of the SI-IQ index (loading k = 0.68, chance level 25%).

Evaluator
OpenRouter
Benchmark home page
openrouter.ai/benchmarks/gpqa-diamond

Results

Source
1 Gemini 4 Argon Google DeepMind96.5%140137–143 8 Sep 2026OpenRouter benchmark datasets
2 Gemini 3.1 Pro Google DeepMind94.8%127124–130 9 Feb 2026OpenRouter benchmark datasets
3 Gemini 3.8 Flash Google DeepMind94.8%130127–133 8 Sep 2026OpenRouter benchmark datasets
4 GPT-6 Astra OpenAI94.7%150146–153 8 Sep 2026OpenRouter benchmark datasets
5 GPT-6.1 Sol OpenAI94.4%148144–152 8 Sep 2026OpenRouter benchmark datasets
6 Gemini 3.7 Flash Google DeepMind93.9%131128–134 8 Aug 2026OpenRouter benchmark datasets
7 GPT-5.5 OpenAI93.6%138135–141 9 Apr 2026OpenRouter benchmark datasets
8 Gemini 3.6 Flash Google DeepMind93.0%125122–128 8 Jul 2026OpenRouter benchmark datasets
9 Claude Sonnet 5.5 Anthropic92.9%140137–143 8 Sep 2026OpenRouter benchmark datasets
10 Grok 4.6 xAI92.5%126123–129 8 Aug 2026OpenRouter benchmark datasets
11 Gemini 3.5 Flash Google DeepMind92.4%126123–129 9 May 2026OpenRouter benchmark datasets
12 GPT-5.6 Sol OpenAI92.3%140137–143 8 Jul 2026OpenRouter benchmark datasets
13 GPT-6 Sol OpenAI92.1%140137–144 8 Sep 2026OpenRouter benchmark datasets
14 Kimi K3open Moonshot91.6%127124–130 8 Jul 2026OpenRouter benchmark datasets
15 MiniMax-M3open MiniMax91.5%108102–114 5 Jun 2026OpenRouter benchmark datasets
16 GLM-5.3-Flashopen Z.ai (Zhipu AI)90.9%120117–124 7 Aug 2026OpenRouter benchmark datasets
17 Qwen3.7-Max Alibaba90.9%125122–128 9 May 2026OpenRouter benchmark datasets
18 Muse Spark Openopen Meta AI90.6%127124–130 6 Oct 2026OpenRouter benchmark datasets
19 Claude Opus 5.5 Anthropic90.2%144141–148 8 Sep 2026OpenRouter benchmark datasets
20 DeepSeek V4 Flash 0731open DeepSeek90.0%124121–127 8 Jul 2026OpenRouter benchmark datasets
21 GPT-5.4 OpenAI90.0%131128–133 9 Mar 2026OpenRouter benchmark datasets
22 DeepSeek-V4-Flashopen DeepSeek89.9%129116–143 2 too little data Apr 2026OpenRouter benchmark datasets
23 MiMo-V2.6 Pro Xiaomi Corp89.9%131117–144 2 too little data Sep 2026OpenRouter benchmark datasets
24 Tencent Hy4 previewopen Tencent89.7%129115–143 2 too little data Aug 2026OpenRouter benchmark datasets
25 Qwen3.8 Flash Alibaba89.6%125111–138 2 too little data Aug 2026OpenRouter benchmark datasets
26 Claude Opus 4.8 Anthropic89.6%132129–135 8 May 2026OpenRouter benchmark datasets
27 Claude Opus 5 Anthropic89.6%136133–140 8 Jul 2026OpenRouter benchmark datasets
28 GPT-5.6 Terra OpenAI89.3%134131–137 8 Jul 2026OpenRouter benchmark datasets
29 DeepSeek V4 Pro 0813open DeepSeek89.3%128125–131 8 Aug 2026OpenRouter benchmark datasets
30 Claude Opus 4.7 Anthropic89.2%127124–130 9 Apr 2026OpenRouter benchmark datasets
31 Claude Fable 5.1 Anthropic88.9%140136–143 8 Sep 2026OpenRouter benchmark datasets
32 Gemini 3 Flash Google DeepMind88.6%122119–125 9 Dec 2025OpenRouter benchmark datasets
33 Kimi K2.7 Codeopen Moonshot88.5%118115–122 7 Jun 2026OpenRouter benchmark datasets
34 DeepSeek V4.1 Flashopen DeepSeek88.4%128125–131 8 Sep 2026OpenRouter benchmark datasets
35 GPT-6 Luna OpenAI88.2%130127–133 8 Sep 2026OpenRouter benchmark datasets
36 Kimi K2.6open Moonshot88.0%121118–124 9 Apr 2026OpenRouter benchmark datasets
37 GPT-5.2 OpenAI87.9%125122–128 10 Dec 2025OpenRouter benchmark datasets
38 Qwen3.7-Plus Alibaba87.4%115110–119 5 Jun 2026OpenRouter benchmark datasets
39 Qwen3.5 397B-A17Bopen Alibaba87.3%114109–118 5 Feb 2026OpenRouter benchmark datasets
40 GPT-5.6 Luna OpenAI87.2%130127–133 8 Jul 2026OpenRouter benchmark datasets
41 Inkling-Smallopen Thinking Machines87.0%118115–122 7 Jul 2026OpenRouter benchmark datasets
42 Claude Haiku 5.5 Anthropic86.9%129126–132 8 Oct 2026OpenRouter benchmark datasets
43 GPT-5.1 OpenAI86.9%120115–125 7 Nov 2025OpenRouter benchmark datasets
44 MiniMax-M2.1open MiniMax86.9%120106–133 2 too little data Dec 2025OpenRouter benchmark datasets
45 GLM-5.3open Z.ai (Zhipu AI)86.4%125122–128 8 Aug 2026OpenRouter benchmark datasets
46 Claude Fable 5 Anthropic86.2%138134–141 8 Jun 2026OpenRouter benchmark datasets
47 Claude Opus 4.5 Anthropic86.1%115111–118 10 Nov 2025OpenRouter benchmark datasets
48 Kimi K2.5open Moonshot85.9%119112–125 5 Jan 2026OpenRouter benchmark datasets
49 Claude Sonnet 5 Anthropic85.8%126123–129 8 Jun 2026OpenRouter benchmark datasets
50 MiniMax-M2.7open MiniMax85.7%120107–134 2 too little data Mar 2026OpenRouter benchmark datasets
51 Claude Opus 4.6 Anthropic85.7%123120–126 9 Feb 2026OpenRouter benchmark datasets
52 MiMo-V2.5-Proopen Xiaomi Corp85.5%121108–135 2 too little data Apr 2026OpenRouter benchmark datasets
53 MiniMax-M2.5open MiniMax85.4%118105–132 2 too little data Feb 2026OpenRouter benchmark datasets
54 DeepSeek-V4-Proopen DeepSeek85.4%117113–120 9 Apr 2026OpenRouter benchmark datasets
55 GPT-5 OpenAI85.0%121118–124 11 Aug 2025OpenRouter benchmark datasets
56 GPT-5.4 Mini OpenAI84.0%117113–120 8 Mar 2026OpenRouter benchmark datasets
57 Gemini 3.5 Flash-Lite Google DeepMind84.0%109106–113 7 Jul 2026OpenRouter benchmark datasets
58 GLM-4.7open Z.ai (Zhipu AI)83.6%112107–118 5 Dec 2025OpenRouter benchmark datasets
59 Qwen 3.8 27Bopen Alibaba83.4%120106–133 2 too little data Aug 2026OpenRouter benchmark datasets
60 GLM-5.2open Z.ai (Zhipu AI)83.3%121118–124 9 Jun 2026OpenRouter benchmark datasets
61 Qwen 3.6 35B-A3Bopen Alibaba82.7%111107–115 5 Apr 2026OpenRouter benchmark datasets
62 DeepSeek-V3.2-Expopen DeepSeek82.7%112103–121 4 too little data Sep 2025OpenRouter benchmark datasets
63 MiMo-V2.5open Xiaomi Corp82.4%117104–131 2 too little data Apr 2026OpenRouter benchmark datasets
64 Inklingopen Thinking Machines81.8%114110–117 7 Jul 2026OpenRouter benchmark datasets
65 Gemma 4 31B ITopen Google DeepMind81.6%10498–110 4 too little data Apr 2026OpenRouter benchmark datasets
66 GLM-5.1open Z.ai (Zhipu AI)81.2%116112–120 7 Apr 2026OpenRouter benchmark datasets
67 Claude Sonnet 4.5 Anthropic81.1%110107–113 11 Sep 2025OpenRouter benchmark datasets
68 Qwen3.5-122B-A10Bopen Alibaba80.7%112103–121 3 too little data Feb 2026OpenRouter benchmark datasets
69 Gemini 2.5 Pro (Jun 2025) Google DeepMind80.6%108104–112 7 Jun 2025OpenRouter benchmark datasets
70 DeepSeek-V3.2open DeepSeek80.3%114101–127 2 too little data Dec 2025OpenRouter benchmark datasets
71 Qwen3.6 27Bopen Alibaba80.1%115110–119 5 Apr 2026OpenRouter benchmark datasets
72 GPT-5 mini OpenAI80.1%115112–118 11 Aug 2025OpenRouter benchmark datasets
73 Kimi K2 Thinkingopen Moonshot79.7%11299–126 2 too little data Nov 2025OpenRouter benchmark datasets
74 Qwen3-235B-A22B-Thinking (Jul 2025)open Alibaba79.5%111105–118 5 Jul 2025OpenRouter benchmark datasets
75 DeepSeek-V3.1-Terminusopen DeepSeek79.4%10289–115 2 too little data Sep 2025OpenRouter benchmark datasets
76 DeepSeek-R1 (May 2025)open DeepSeek78.5%104100–109 4 too little data May 2025OpenRouter benchmark datasets
77 Gemma 4 26B A4Bopen Google DeepMind77.6%106100–112 3 too little data Apr 2026OpenRouter benchmark datasets
78 GPT-5.4 Nano OpenAI77.6%118115–121 8 Mar 2026OpenRouter benchmark datasets
79 DeepSeek-V3.1open DeepSeek77.1%10087–113 2 too little data Aug 2025OpenRouter benchmark datasets
80 Claude Sonnet 4 Anthropic76.7%9793–101 5 May 2025OpenRouter benchmark datasets
81 GLM-4.6open Z.ai (Zhipu AI), Tsinghua University76.5%117106–128 3 too little data Sep 2025OpenRouter benchmark datasets
82 MiMo-V2.6 Flash Xiaomi Corp75.9%11299–125 2 too little data Sep 2026OpenRouter benchmark datasets
83 gpt-oss-120bopen OpenAI75.8%10699–112 5 Aug 2025OpenRouter benchmark datasets
84 qwen3-coder-next —75.8%10390–116 2 too little data Feb 2026OpenRouter benchmark datasets
85 Mistral Small 4open Mistral AI75.3%9683–108 2 too little data Mar 2026OpenRouter benchmark datasets
86 Qwen3.5-35B-A3Bopen Alibaba75.0%106100–112 3 too little data Feb 2026OpenRouter benchmark datasets
87 Kimi K2 (Sep 2025)open Moonshot74.7%10592–118 2 too little data Sep 2025OpenRouter benchmark datasets
88 Qwen3.5-9Bopen Alibaba74.6%10195–107 3 too little data Feb 2026OpenRouter benchmark datasets
89 Qwen3-235B-A22B-Instruct (Jul 2025)open Alibaba73.9%9685–107 3 too little data Jul 2025OpenRouter benchmark datasets
90 Gemini 2.5 Flash (Jun 2025) Google DeepMind73.8%10291–113 3 too little data Jun 2025OpenRouter benchmark datasets
91 Claude Haiku 4.5 Anthropic73.1%10499–108 7 Oct 2025OpenRouter benchmark datasets
92 GLM-5open Z.ai (Zhipu AI)72.9%112105–118 4 too little data Feb 2026OpenRouter benchmark datasets
93 GPT-5 nano OpenAI68.4%105102–109 10 Aug 2025OpenRouter benchmark datasets
94 DeepSeek-V3open DeepSeek66.2%7773–82 4 too little data Dec 2024OpenRouter benchmark datasets
95 GPT-4.1 OpenAI66.1%9187–94 10 Apr 2025OpenRouter benchmark datasets
96 Llama 4 Maverickopen Meta AI64.8%8177–85 6 Apr 2025OpenRouter benchmark datasets
97 GPT-4.1 mini OpenAI64.7%9188–95 10 Apr 2025OpenRouter benchmark datasets
98 DeepSeek-V3 (Mar 2025)open DeepSeek64.6%8581–90 4 too little data Mar 2025OpenRouter benchmark datasets
99 Claude Sonnet 4.6 Anthropic64.6%113108–118 7 Feb 2026OpenRouter benchmark datasets
100 Qwen3-30B-A3B-Instruct (Jul 2025)open Alibaba64.2%9186–97 4 too little data Jul 2025OpenRouter benchmark datasets
101 gpt-oss-20bopen OpenAI63.1%9589–101 3 too little data Aug 2025OpenRouter benchmark datasets
102 Qwen3-30B-A3Bopen Alibaba60.4%9488–100 3 too little data Apr 2025OpenRouter benchmark datasets
103 Qwen3-Coder-480B-A35Bopen Alibaba60.4%8572–98 2 too little data Jul 2025OpenRouter benchmark datasets
104 Qwen3-32Bopen Alibaba55.7%9589–100 5 Apr 2025OpenRouter benchmark datasets
105 GPT-4o (May 2024) OpenAI54.2%7066–75 4 too little data May 2024OpenRouter benchmark datasets
106 Qwen3-14Bopen Alibaba53.3%9387–99 4 too little data Apr 2025OpenRouter benchmark datasets
107 GPT-4o (Aug 2024) OpenAI52.8%7368–77 8 Aug 2024OpenRouter benchmark datasets
108 GPT-4.1 nano OpenAI49.7%7672–80 8 Apr 2025OpenRouter benchmark datasets
109 glm-4.7-flash —49.2%9488–100 3 too little data —OpenRouter benchmark datasets
110 Llama 3.3 70Bopen Meta AI47.5%6661–71 5 Dec 2024OpenRouter benchmark datasets
111 Qwen2.5-72Bopen Alibaba46.1%7268–77 4 too little data Sep 2024OpenRouter benchmark datasets
112 GPT-4o mini OpenAI43.1%6864–73 9 Jul 2024OpenRouter benchmark datasets
113 Gemma 3 12Bopen Google DeepMind37.9%7366–79 4 too little data Mar 2025OpenRouter benchmark datasets
114 Qwen2.5-7Bopen Alibaba34.2%5141–60 3 too little data Sep 2024OpenRouter benchmark datasets
115 Gemma 3 4Bopen Google DeepMind27.1%6353–73 3 too little data Mar 2025OpenRouter benchmark datasets
116 Llama 3.1-8Bopen Meta AI26.9%5549–60 3 too little data Jul 2024OpenRouter benchmark datasets
117 Llama 3.2 3Bopen Meta AI17.3%6142–80 2 too little data Sep 2024OpenRouter benchmark datasets

Data sources: Aider polyglot leaderboard data (Apache-2.0); Arena leaderboard dataset on Hugging Face (CC BY 4.0); Berkeley Function Calling Leaderboard, Gorilla project (Apache-2.0); Epoch AI, ‘Capabilities & benchmarking’ (CC BY 4.0); API list prices via LiteLLM's price table (MIT); each price links to the developer's pricing page; Source: OpenRouter (openrouter.ai), licensed under CC BY 4.0: reuse and republish with attribution to OpenRouter..