PrimeIntellect-ai

How the prime-rl config system works — TOML files, CLI overrides, composition, and special patterns. Use when creating configs, debugging config errors, or overriding values via CLI.

Monitor an ongoing prime-rl training run — find the output directory, tail logs, check key metrics, inspect SLURM jobs, and restart safely. Use when asked to check on a run, debug training, or investigate performance.

Launch and monitor prime-rl training runs. Use when starting, supervising, or debugging an RL/SFT run. Routes to `start-run` (entrypoints + how to launch) and `monitor-run` (logs, metrics, check-ins).

How to install prime-rl and its optional dependencies. Use when setting up the project, installing extras like DeepEP for multi-node expert parallelism, or troubleshooting dependency issues.

How to launch prime-rl training runs — the `rl`, `sft`, and `inference` entrypoints, their config classes, and single-node/SLURM/dry-run modes. Use when starting a run or picking the right entrypoint.

How to prepare and publish GitHub releases for prime-rl. Use when drafting release notes, tagging versions, or publishing releases.

Create or migrate native verifiers.v1 taskset, environment, and harness packages. Use to build a taskset, port a benchmark, add task tools or user simulation, build a multi-agent environment, package an agent harness, or migrate an existing v0 environment to the typed v1 trace model.

Run and evaluate verifiers tasksets. Set up the necessary config files and observe the runs and their results.

Run interactive brainstorming across verifiers tasksets, evaluations, GEPA, and RL training. Use when the user wants ideation, literature scanning, concept teaching, roadmap planning, or research program design grounded in local CLI sources, verifiers, and RL trainer code.

Install and run a verifiers environment — smoke testing during development and full benchmark evals. Covers the `eval` CLI for the `*_v1` tasksets. Use while developing/iterating on an environment, when running/evaluating/benchmarking one, or when the user mentions eval, tasksets, eval runs, or smoke tests.