toml-config

Every prime-rl command takes its settings the same way, and this is that system — the file format, its traps, and how the pieces stack on top of each other.

WHEN TO USE

SCOPE

USE FOR

Write a TOML config file for a prime-rl command such as training or inference.
Override a value from a config file directly on the command line.
Combine multiple TOML config files, with later files overriding earlier ones.
Set a config value to None even though plain TOML has no null value.
Submit an RL or SFT training run to a shared cluster by adding a slurm section.
Look up which command to run for a given task, such as training or serving.

This is a saved copy of the skill's SKILL.md instructions. It may differ from the upstream file or the version you install.

In this document (7)

TOML Config

All prime-rl commands use pydantic_config (tyro-backed) with TOML configs and CLI overrides.

Running with configs

# Load a config file with @ syntax
uv run inference @ configs/debug/infer.toml
uv run sft @ configs/debug/sft/train.toml
uv run rl @ configs/debug/rl/train.toml

# CLI overrides (take precedence over TOML)
uv run inference @ config.toml --model.name Qwen/Qwen3-0.6B --server.port 8001

# Boolean flags: no value needed
uv run inference --model.enforce-eager          # sets to true
uv run inference --no-model.enforce-eager       # sets to false

# CLI-only (no TOML file)
uv run inference --model.name Qwen/Qwen3-0.6B --model.max-model-len 2048

# Compose multiple config files (later files override earlier ones)
uv run rl @ examples/reverse_text/rl.toml @ examples/reverse_text/slurm_rl.toml

# Nested config files: load a config for a specific section
uv run rl --model @ model.toml --data @ data.toml

TOML structure

Top-level fields must come before any [section] header — this is a TOML rule.

# Top-level fields first
gpu_memory_utilization = 0.5
seed = 42

# Then sections
[model]
name = "Qwen/Qwen3-0.6B"
max_model_len = 4096

[server]
port = 8000

Putting a top-level field after a section header nests it inside that section, which causes validation errors.

Setting None

Use the string "None" in TOML to set a field to None:

max_model_len = "None"

SLURM mode

Both rl and sft commands support SLURM execution via an optional [slurm] section. When present, the run is submitted as a SLURM job instead of running locally.

SLURM configs are composed with the base config via CLI:

uv run rl @ examples/reverse_text/rl.toml @ examples/reverse_text/slurm_rl.toml

RL SLURM

output_dir = "/shared/experiments/my-run"

[deployment]
type = "multi_node"
num_train_nodes = 2
num_infer_nodes = 1
gpus_per_node = 8
# nodes_per_fsdp_group = 1

[slurm]
job_name = "my-rl-job"
# dry_run = true          # generate script without submitting
# template_path = "path/to/custom.sh.j2"
# project_dir = "/path/to/project"

When [slurm] is set for RL:

output_dir must be explicitly set (the default outputs is rejected)
Teacher inference is not supported in multi-node deployment

SFT SLURM

output_dir = "/shared/experiments/my-sft-run"

[deployment]
type = "multi_node"
num_nodes = 2
gpus_per_node = 8
# nodes_per_fsdp_group = 1

[slurm]
job_name = "my-sft-job"
# dry_run = true
# template_path = "path/to/custom.sh.j2"
# project_dir = "/path/to/project"

SFT deployment follows the same pattern as RL:

[deployment] configures node/GPU allocation (single_node default or multi_node)
[slurm] configures SLURM submission (job name, partition, template)
output_dir must be explicitly set when using SLURM
Multi-node deployment requires [slurm] to be set

Available commands

All accept @ config.toml and CLI overrides:

Command	Config class	Description
`uv run rl`	full RL pipeline	Orchestrator + inference + trainer (local or SLURM)
`uv run inference`	`InferenceConfig`	vLLM inference server
`uv run trainer`	trainer config	RL trainer
`uv run orchestrator`	orchestrator config	Rollout orchestrator
`uv run env-server`	env server config	Environment server
`uv run sft`	SFT config	Supervised fine-tuning (local or SLURM)

Key files

src/prime_rl/utils/config.py — BaseConfig, cli, get_all_fields
src/prime_rl/entrypoints/rl.py — unified RL entrypoint (local + SLURM)
src/prime_rl/configs/rl.py — RLConfig, SlurmConfig, DeploymentConfig
src/prime_rl/entrypoints/sft.py — unified SFT entrypoint (local + SLURM)
src/prime_rl/configs/sft.py — SFTConfig
configs/ — all config files, organized by task

Local tool required

1 FOR YOU

01
Install prime-rl and its dependencies
Run uv sync to install prime-rl before running any of its commands (rl, sft, inference, etc.) with a TOML config.
```
uv sync
```

Listed as free: Connected services may require an account or charge fees.

Not standalone. This skill ships with huggingface/prime-rl and only works together with that tool — install the tool first, then add this skill.

Run this outside your AI tool, using Node.js 22.20.0 or newer. Skills CLI 1.7.2 skips skill selection; prompts cancel installation without an interactive terminal. If choosing Project when asked about installation scope, open your project's folder before running this.

Install checklist

Check selected tools if a picker appears; Claude Code may be unchecked under Additional agents. ↑/↓ moves, Space toggles selections, Enter confirms.
Without a tool picker, check Installing to: for your intended tool. Stop if it names a different tool.
If asked, choose Symlink (Recommended) for shared files or Copy to all agents for separate copies; review the summary before confirming installation.
Install the find-skills skill? defaults to Yes, adding another skill for all projects. Choose No if you don't want that extra installation.

Skills CLI 1.7.2 sends install data to skills.sh by default, including the source, skill names, target tools and file details; GitHub sources are reported only when GitHub confirms the repository is public, while other remote sources can also be reported. Confirmed-public GitHub sources also trigger a security-audit request. Before installing, set the environment variable DO_NOT_TRACK=1 or DISABLE_TELEMETRY=1 to disable both.

npx skills add huggingface/prime-rl --skill "toml-config" --full-depth

After installing for Claude Code, open it in the installation project if you chose Project; Global makes the skill available across projects on this computer. Once Claude Code loads the skill, its description helps Claude decide when to use it.

If installation created a top-level skills folder that did not exist when your Claude Code session started, run /reload-skills in that session. Run it again after later changes in that folder.

What's inside

Files included with this skill.