diff --git a/README.md b/README.md index 407a5e9..22394fc 100644 --- a/README.md +++ b/README.md @@ -355,6 +355,22 @@ PR-AF is one example of a larger AgentField ecosystem for building autonomous, a * **[SWE-AF](http://www.agentfield.ai/github/swe-af)**: autonomous software factory for production ready PR * **[SEC-AF](http://www.agentfield.ai/github/sec-af)**: autonomous security factory for code security review, vulnerability investigation, and evidence-grounded remediation. +## New from AgentField: CodeAF + + + First on DeepSWE: senior-dev, CodeAF's developer subharness, solved the most tasks (54.9%) at the lowest cost per solved task (1x). Every other harness solved less and paid more per solve: mini-swe-agent 1.9x, codex 2.1x, pi 2.4x, claude-code 3.4x, omp, kilo and opencode about 4.5x, muse-code 11.3x, deepseek-harness 26.6x. + + +**[CodeAF](https://agentfield.ai/github/codeaf/?utm_source=github-readme&utm_campaign=pr-af-readme&utm_id=pr-af-readme-codeaf-repo)** is our new open-source coding harness, +built for open models: frontier-grade coding at a fraction of the cost. It ranked +**#1 on DeepSWE** out of ten coding harnesses on the same model (DeepSeek V4 Flash), +ahead of Claude Code, Codex and OpenCode, at the lowest cost per solved issue. +One Go binary, Apache 2.0. + +PR-AF also ships inside CodeAF as its `review` subharness. + +[Try CodeAF →](https://agentfield.ai/github/codeaf/?utm_source=github-readme&utm_campaign=pr-af-readme&utm_id=pr-af-readme-codeaf-repo) · [See the benchmark →](https://agentfield.ai/github/codeaf/?utm_source=github-readme&utm_campaign=pr-af-readme&utm_id=pr-af-readme-codeaf-benchmark#benchmarks) + ## From the AgentField Blog ### [How an AI-Native Engineering Team Does Code Review](https://www.agentfield.ai/blog/ai-native-code-review?utm_source=github-readme&utm_campaign=pr-af-readme&utm_id=pr-af-readme-blog-ai-native-code-review)