piwheels

coder-eval

Evaluate, benchmark, and A/B-test AI coding agents (Claude Code, Codex, Gemini/Antigravity) with sandboxed, reproducible YAML task suites.

Installation

In a virtualenv (see these instructions if you need to create one):

pip3 install coder-eval

Dependencies

• anyio
• pydantic-settings
• anthropic
• python-dotenv
• rich
• azure-monitor-opentelemetry-exporter
• claude-agent-sdk
• tqdm
• click
• opentelemetry-sdk
• pydantic
• typer
• jmespath
• jsonschema
• starlette
• pyyaml
• radon

Releases

Version	Released	Bullseye Python 3.9	Bookworm Python 3.11	Trixie Python 3.13	Files
0.8.10	2026-07-24
0.8.9	2026-07-23
0.8.8	2026-07-22
0.8.7	2026-07-22
0.8.7rc11 pre-release	2026-07-21
0.8.7rc10 pre-release	2026-07-21
0.8.6	2026-07-20
0.8.5	2026-07-20
0.8.4	2026-07-13
coder_eval-0.8.4-py3-none-any.whl (479 KB)
0.8.3	2026-07-09
coder_eval-0.8.3-py3-none-any.whl (465 KB)
0.8.2	2026-07-09
coder_eval-0.8.2-py3-none-any.whl (465 KB)

Issues with this package?

• Search issues for this package
• Package or version missing? Open a new issue
• Something else? Open a new issue