This is an early release preview. You may encounter bugs.
rLLM logo
Unclaimed

Tool training qa

rLLM

rLLM bolts RL onto agents you already wrote. verl, trlx, OpenRLHF make you rewrite the agent into their pipeline. A deco

A 84/100 GitHub score ? This grade is derived from GitHub signals, not user votes. Open for the full breakdown.
No votes yet

01 / About

What rLLM is.

rLLM trains language agents with reinforcement learning without rewriting them. You point it at an agent you already have, and the same agent code drives both evaluation and training: a model gateway sits in front of the agent's LLM calls, captures token IDs and log probabilities by URL-routed session, and structures them into episodes (one task), trajectories (one agent run), and steps (one LLM call). A reward function scores the result and the RL algorithm updates the model weights.

Agents come from two directions. Over ten command-line harnesses are supported directly, including Claude Code, Codex, Terminus-2, mini-swe-agent, and opencode, alongside Harbor-compatible task directories. Any other program — a LangGraph graph, an OpenAI Agents SDK agent, or a plain openai.OpenAI call — is wrapped with the @rllm.rollout decorator. Rollouts run in Docker, Daytona, Modal, or locally, with snapshotting and warm pools to keep training-scale rollouts cheap.

Training backends are selected with a single flag: verl for distributed multi-GPU runs with vLLM or SGLang, tinker for a single machine, and fireworks for the Fireworks platform. Supported methods include GRPO, REINFORCE, RLOO, supervised fine-tuning, and on-policy distillation. Over 60 benchmarks are integrated across maths, code, multiple choice, question answering, search, vision-language, translation, and agentic tasks — Terminal-Bench 2.0, SWE-bench, SkillsBench, AIME, MATH-500, and GPQA among them — and rllm eval pulls and runs one.

There are two entry points: a CLI (rllm model setup, rllm eval, rllm train) that needs no code, and a Python API where you decorate a rollout function and an evaluator function and hand them to an AgentTrainer. The project reports open-source results for DeepScaleR-1.5B, DeepCoder-14B, DeepSWE-32B, and FinQA-4B, and lists community projects built on it, including Tongyi DeepResearch, Terminal-Bench-RL, PettingLLMs, and SETA. It requires Python 3.11 or later.

Features

  • Any harness: 10+ CLI harnesses including Claude Code, Codex, Terminus-2, mini-swe-agent, and opencode, plus Harbor-compatible task directories
  • Wrap your own agent: the @rllm.rollout decorator turns existing LangGraph, OpenAI Agents SDK, or raw OpenAI-client code into a trainable rollout
  • Any sandbox: Docker, Daytona, Modal, or local execution, with snapshot and warm-pool acceleration
  • Swappable backends: verl for distributed multi-GPU, tinker for single-machine, and fireworks, selected by one flag
  • Training methods: GRPO, REINFORCE, RLOO, supervised fine-tuning, and on-policy distillation
  • 60+ benchmarks: Terminal-Bench 2.0, SWE-bench, SkillsBench, AIME, MATH-500, GPQA, and more, run with rllm eval
  • Model gateway: captures token IDs and log probabilities behind the scenes so eval and training share one agent implementation
  • CLI or Python API: rllm model setup, rllm eval, and rllm train, or a rollout plus evaluator handed to AgentTrainer

02 / Discussion CREDIBILITY-GATED

Discussion

Reading is open to everyone. Posting and voting need a verified identity or a GitHub grade of B or higher.

  • No discussions yet.

04 / Build

Build with rLLM.

Browse the catalogue for frameworks, tools, and harnesses, each scored on real GitHub credibility.

Get rLLM →

Browse the catalogue