Which AI model scores highest on FrontierCode 2.0?

Claude Fable 5.2 by Anthropic holds the best FrontierCode 2.0 result among tracked models, at 49.1% (released Nov 9 2026). Higher scores are better on this benchmark.

What are the top 5 models on FrontierCode 2.0?

1. Claude Fable 5.2 (Anthropic) — 49.1%; 2. Claude Opus 5.5 (Anthropic) — 45.5%; 3. GPT-6 Astra (OpenAI) — 42.8%; 4. Gemini 4 Carbon (Google) — 41.1%; 5. Claude Fable 5.1 (Anthropic) — 32.4%.

How many models have a published FrontierCode 2.0 score?

5 tracked models have a published FrontierCode 2.0 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.

Agentic coding

FrontierCode 2.0

whether a model’s code changes to open-source repositories would be merged by their maintainers (version 2.0); separate from FrontierCode v1.1 scores

Rankings

Higher is better

FrontierCode 2.0 — frequently asked questions

: whether a model’s code changes to open-source repositories would be merged by their maintainers (version 2.0); separate from FrontierCode v1.1 scores
: Claude Fable 5.2 by Anthropic holds the best FrontierCode 2.0 result among tracked models, at 49.1% (released Nov 9 2026). Higher scores are better on this benchmark.
: 1. Claude Fable 5.2 (Anthropic) — 49.1%; 2. Claude Opus 5.5 (Anthropic) — 45.5%; 3. GPT-6 Astra (OpenAI) — 42.8%; 4. Gemini 4 Carbon (Google) — 41.1%; 5. Claude Fable 5.1 (Anthropic) — 32.4%.
: 5 tracked models have a published FrontierCode 2.0 score. Scores are collected from published lab reports and benchmark sources, with recorded sources credited alongside the results. We do not re-run the evaluations.