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

Aider Polyglot: AI model results

Aider Polyglot is a Coding benchmark run by Aider. 40 models in our data have a published result.

It is part of the SI-IQ index (loading k = 1.06, chance level 0%).

Evaluator
Aider
Benchmark home page
aider.chat/docs/leaderboards

Results

Source
1 GPT-5 OpenAI88.0%121118–124 11 Aug 2025Aider LLM Leaderboards
2 o3-pro OpenAI84.9%119103–135 1 too little data Jun 2025Aider LLM Leaderboards
3 Gemini 2.5 Pro (Jun 2025) Google DeepMind83.1%108104–112 7 Jun 2025Aider LLM Leaderboards
4 o3 OpenAI81.3%116113–119 10 Apr 2025Aider LLM Leaderboards
5 Grok 4 xAI79.6%114107–120 4 too little data Jul 2025Aider LLM Leaderboards
6 Gemini 2.5 Pro (May 2025) Google DeepMind76.9%10498–110 3 too little data May 2025Aider LLM Leaderboards
7 DeepSeek-V3.2-Expopen DeepSeek74.2%112103–121 4 too little data Sep 2025Aider LLM Leaderboards
8 Gemini 2.5 Pro (Mar 2025) Google DeepMind72.9%108102–113 3 too little data Mar 2025Aider LLM Leaderboards
9 Claude Opus 4 Anthropic72.0%9893–102 5 May 2025Aider LLM Leaderboards
10 o4-mini OpenAI72.0%112109–115 10 Apr 2025Aider LLM Leaderboards
11 Claude 3.7 Sonnet Anthropic64.9%9995–103 5 Feb 2025Aider LLM Leaderboards
12 Claude Sonnet 4 Anthropic61.3%9793–101 5 May 2025Aider LLM Leaderboards
13 Qwen3-235B-A22Bopen Alibaba59.6%8580–91 3 too little data Apr 2025Aider LLM Leaderboards
14 DeepSeek-R1open DeepSeek56.9%9893–102 4 too little data Jan 2025Aider LLM Leaderboards
15 Gemini 2.5 Flash (May 2025) Google DeepMind55.1%10294–109 2 too little data May 2025Aider LLM Leaderboards
16 o3-mini OpenAI53.8%105101–108 9 Jan 2025Aider LLM Leaderboards
17 Grok 3 xAI53.3%9691–100 4 too little data Apr 2025Aider LLM Leaderboards
18 GPT-4.1 OpenAI52.4%9187–94 10 Apr 2025Aider LLM Leaderboards
19 Claude 3.5 Sonnet (October 2024) Anthropic51.6%7671–80 4 too little data Oct 2024Aider LLM Leaderboards
20 Grok-3 mini xAI49.3%10095–105 4 too little data Apr 2025Aider LLM Leaderboards
21 Gemini 2.5 Flash (Apr 2025) Google DeepMind47.1%10194–108 2 too little data Apr 2025Aider LLM Leaderboards
22 gpt-oss-120bopen OpenAI41.8%10699–112 5 Aug 2025Aider LLM Leaderboards
23 Qwen3-32Bopen Alibaba40.0%9589–100 5 Apr 2025Aider LLM Leaderboards
24 Gemini 2.0 Pro Google DeepMind35.6%8983–94 3 too little data Feb 2025Aider LLM Leaderboards
25 o1-mini OpenAI32.9%9287–96 4 too little data Sep 2024Aider LLM Leaderboards
26 GPT-4.1 mini OpenAI32.4%9188–95 10 Apr 2025Aider LLM Leaderboards
27 Claude 3.5 Haiku Anthropic28.0%6762–72 4 too little data Oct 2024Aider LLM Leaderboards
28 GPT-4o (Aug 2024) OpenAI23.1%7368–77 8 Aug 2024Aider LLM Leaderboards
29 Gemini 2.0 Flash (Dec 2024) Google DeepMind, Google22.2%8064–95 1 too little data Dec 2024Aider LLM Leaderboards
30 Qwen2.5-Max Alibaba21.8%7873–82 4 too little data Jan 2025Aider LLM Leaderboards
31 QwQ-32Bopen Alibaba20.9%9286–98 3 too little data Mar 2025Aider LLM Leaderboards
32 GPT-4o (Nov 2024) OpenAI18.2%6964–73 5 Nov 2024Aider LLM Leaderboards
33 Gemini 2.0 Flash Thinking (Jan 2025) Google DeepMind, Google18.2%8983–95 3 too little data Jan 2025Aider LLM Leaderboards
34 Qwen2.5-Coder-32B-Instructopen Alibaba16.4%7559–90 1 too little data Nov 2024Aider LLM Leaderboards
35 Llama 4 Maverickopen Meta AI15.6%8177–85 6 Apr 2025Aider LLM Leaderboards
36 Yi-Lightning 01.AI12.9%7155–87 1 too little data Dec 2024Aider LLM Leaderboards
37 openhands-lm-32b-v0.1 —10.2%6752–83 1 too little data Mar 2024Aider LLM Leaderboards
38 GPT-4.1 nano OpenAI8.9%7672–80 8 Apr 2025Aider LLM Leaderboards
39 Gemma 3 27Bopen Google DeepMind4.9%7772–81 5 Mar 2025Aider LLM Leaderboards
40 GPT-4o mini OpenAI3.6%6864–73 9 Jul 2024Aider LLM Leaderboards

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..