// BENCHMARK

Aider Polyglot benchmark

AI model leaderboard for the Aider Polyglot benchmark. Compare how large language models score on Aider Polyglot, see the full ranking, and understand what this AI benchmark measures. Claude Opus 4.7 currently leads with 87%. Aider editor benchmark covering Python, JavaScript, Rust, Go, C++, and more — measures real-world code-editing skill.

Leaderboard

# Model Organization Score Variant Source
#1 Claude Opus 4.7 Anthropic 87% official ↗
#2 DeepSeek-V3.2-Exp DeepSeek 74.5% from-hf-readme official ↗
#3 DeepSeek-R1 DeepSeek 53.3% cited-magistral official ↗
#4 DeepSeek-V3 DeepSeek 49.6% cited-magistral official ↗
#5 Magistral Medium 1.1 Mistral AI 47.1% official ↗
#6 gpt-oss-120b OpenAI 44.4% high reasoning official ↗
#7 gpt-oss-120b OpenAI 44.4% high official ↗
#8 gpt-oss-20b OpenAI 34.2% high reasoning official ↗
#9 gpt-oss-20b OpenAI 34.2% high official ↗
#10 Mistral Medium 3 Mistral AI 28.9% cited-magistral official ↗

Frequently asked questions about Aider Polyglot

What is the Aider Polyglot benchmark?

Aider editor benchmark covering Python, JavaScript, Rust, Go, C++, and more — measures real-world code-editing skill.

How is the Aider Polyglot benchmark scored?

Aider Polyglot is scored using the pass-rate (%) metric, where a higher score is better. The maximum achievable score is 100.000. GenAIList aggregates reported scores from model providers and papers into a single ranked leaderboard.

Which AI model scores highest on Aider Polyglot?

As of the latest reported scores on GenAIList, Claude Opus 4.7 achieves the highest result on Aider Polyglot with a score of 87%.

Is a higher Aider Polyglot score better?

Yes. On Aider Polyglot a higher score indicates better performance, so models near the top of the leaderboard are the strongest.