Pre-release

This release is a pre-release and may not be stable for production use.

Prime Intellect

Verifiers: Environments for LLM Reinforcement Learning

Documentation • Environments Hub • PRIME-RL

News & Updates

[05/07/26] v0.1.14 is released, featuring the v1 Taskset/Harness API, shared eval and training config shape, model-family starter configs, OpenAI Responses and renderer-backed clients, per-turn timing, GEPA prompt artifacts, Lean guard markers, and release/infrastructure hardening.
[04/28/26] v0.1.13.dev8 is released, featuring per-rollout wall-clock timeouts for MultiTurnEnv, CLI timeout config, sandbox timeout propagation, and smaller CliAgentEnv and RLM fixes.
[04/17/26] v0.1.12 is released, featuring upstreamed opencode and RLM harnesses/tasksets, major RLMEnv improvements (context dropping, prompt builder, hardened transport), multi-worker env server support, expanded vf-tui capabilities, and richer eval configuration.
[03/12/26] v0.1.11 is released, featuring a unified client stack, major RLMEnv and env server reliability improvements, a substantially refined eval TUI, new pass@k and ablation sweep support, and bundled opencode environments.
[02/10/26] v0.1.10 is released, featuring OpenEnv and BrowserEnv integrations, resumed evals, improved rollout and token tracking, safer sandbox lifecycle behavior, refreshed workspace setup, and opencode harbor improvements.
[01/08/26] v0.1.9 is released, featuring a number of new experimental environment class types, monitor rubrics for automatic metric collection, improved workspace setup flow, improved error handling, bug fixes, and a documentation overhaul.
[11/19/25] v0.1.8 is released, featuring a major refactor of the rollout system to use trajectory-based tracking for token-in token-out training across turns, as well as support for truncated or branching rollouts.
[11/07/25] Verifiers v0.1.7 is released! This includes an improved quickstart configuration for training with prime-rl, a new included "nano" trainer (vf.RLTrainer, replacing vf.GRPOTrainer), and a number of bug fixes and improvements to the documentation.
[10/27/25] A new iteration of the Prime Intellect Environments Program is live!

Overview

Verifiers is our library for creating environments to train and evaluate LLMs.

Environments contain everything required to run and evaluate a model on a particular task:

A dataset of task inputs
A harness for the model (tools, sandboxes, context management, etc.)
A reward function or rubric to score the model's performance

Environments can be used for training models with reinforcement learning (RL), evaluating capabilities, generating synthetic data, experimenting with agent harnesses, and more.

Verifiers is tightly integrated with the Environments Hub, as well as our training framework prime-rl and our Hosted Training platform.

Getting Started

Ensure you have uv installed, as well as the prime CLI tool:

# install uv
curl -LsSf https://astral.sh/uv/install.sh | sh
# install the prime CLI
uv tool install prime
# log in to the Prime Intellect platform
prime login

To set up a new workspace for developing environments, do:

# ~/dev/my-lab
prime lab setup

This sets up a Python project if needed (with uv init), installs verifiers (with uv add verifiers), creates the recommended workspace structure, and downloads useful starter files:

configs/
├── endpoints.toml      # OpenAI-compatible API endpoint configuration
├── rl/                 # Example configs for Hosted Training
├── eval/               # Example multi-environment eval configs
└── gepa/               # Example configs for prompt optimization
.prime/
└── skills/             # Bundled workflow skills for create/browse/review/eval/GEPA/train/brainstorm
environments/
└── AGENTS.md           # Documentation for AI coding agents
AGENTS.md               # Top-level documentation for AI coding agents
CLAUDE.md               # Claude-specific pointer to AGENTS.md

Alternatively, add verifiers to an existing project:

uv add verifiers && prime lab setup --skip-install

Environments built with Verifiers are self-contained Python modules. To initialize a fresh environment template, do:

prime env init my-env # creates a new template in ./environments/my_env

Add an explicit harness loader when the environment owns harness behavior:

prime env init my-env --with-harness

For OpenEnv integration, use:

prime env init my-openenv --openenv

Then copy your OpenEnv project into environments/my_openenv/proj/ and build the image with:

uv run vf-build my-openenv

This will create a new module called my_env with a basic environment template.

environments/my_env/
├── my_env.py           # Main implementation
├── pyproject.toml      # Dependencies and metadata
└── README.md           # Documentation

Environment modules should expose a load_environment function which returns an environment object. For simple legacy environments, this can still be a direct constructor:

# my_env.py
import verifiers as vf

def load_environment(dataset_name: str = 'gsm8k') -> vf.Environment:
    dataset = vf.load_example_dataset(dataset_name) # 'question'
    async def correct_answer(completion, answer) -> float:
        completion_ans = completion[-1]['content']
        return 1.0 if completion_ans == answer else 0.0
    rubric = vf.Rubric(funcs=[correct_answer])
    env = vf.SingleTurnEnv(dataset=dataset, rubric=rubric)
    return env

For new environments with reusable tasksets, toolsets, custom programs, or custom harnesses, use the v1 Taskset/Harness path:

# my_env.py
import verifiers as vf


class MyTasksetConfig(vf.TasksetConfig):
    system_prompt: vf.SystemPrompt = "Reverse text exactly."


class MyTaskset(vf.Taskset[MyTasksetConfig]):
    def load_tasks(self, split: vf.TaskSplit = "train") -> vf.Tasks:
        rows = [
            {
                "prompt": [{"role": "user", "content": "Reverse abc."}],
                "answer": "cba",
                "split": "train",
                "max_turns": 1,
            }
        ]
        return [row for row in rows if row["split"] == split]

    @vf.reward(weight=1.0)
    async def contains_answer(self, task, state) -> float:
        return float(task["answer"] in str(state.get("completion") or ""))


def load_taskset(config: MyTasksetConfig) -> MyTaskset:
    return MyTaskset(config=config)


def load_environment(config: vf.EnvConfig) -> vf.Env:
    """Loader pattern for all Taskset/Harness environments."""
    return vf.Env(
        taskset=vf.load_taskset(config=config.taskset),
        harness=vf.load_harness(config=config.harness),
    )

The child loader annotation defines the taskset config shape; root load_environment stays typed as vf.EnvConfig. See BYO Harness for the advanced v1 taskset/harness API. Reusable taskset and harness packages live in tasksets and harnesses. Install them with uv add "verifiers[packages]", or with the narrower verifiers[tasksets], verifiers[harnesses], and backend-specific extras. For example, Harbor task directories can run through the bundled OpenCode CLI harness with:

from harnesses import OpenCode, OpenCodeConfig
from tasksets import HarborTaskset, HarborTasksetConfig

env = vf.Env(
    taskset=HarborTaskset(config=HarborTasksetConfig(bundle_package=__name__)),
    harness=OpenCode(config=OpenCodeConfig()),
)

The same environment package is the unit used by evals and prime-rl. The trainer owns model, endpoint, sampling, and rollout count; v1-specific options stay on the taskset or harness config that owns them:

# configs/rl/my-v1-env.toml
model = "Qwen/Qwen3-30B-A3B-Instruct-2507"
max_steps = 100
batch_size = 256
rollouts_per_example = 8

[sampling]
max_tokens = 4096

[[env]]
id = "my-env"

[env.harness]
max_turns = 1

[env.taskset]
system_prompt = "Reverse text exactly."

[env.taskset.scoring.contains_answer]
weight = 1.0

prime env install my-env

For self-managed training launch commands, use the prime-rl documentation.

To run a local evaluation with any OpenAI-compatible model, do:

prime eval run my-env -m openai/gpt-5-nano # run and save eval results locally

Evaluations use Prime Inference by default; configure your own API endpoints in ./configs/endpoints.toml.

View local evaluation results in the terminal UI:

prime eval view

To publish the environment to the Environments Hub, do:

prime env push --path ./environments/my_env

To run an evaluation directly from the Environments Hub, do:

prime eval run primeintellect/math-python

Documentation

Environments — Create datasets, rubrics, and custom multi-turn interaction protocols.

BYO Harness — Build v1 Taskset/Harness environments with custom tools, sandboxes, users, and custom programs.

Evaluation - Evaluate models using your environments.

Training — Train models in your environments with reinforcement learning.

Development — Contributing to verifiers

API Reference — Understanding the API and data structures

FAQs - Other frequently asked questions.

Citation

Originally created by Will Brown (@willccbb).

If you use this code in your research, please cite:

@misc{brown_verifiers_2025,
  author       = {William Brown},
  title        = {{Verifiers}: Environments for LLM Reinforcement Learning},
  howpublished = {\url{https://github.com/PrimeIntellect-ai/verifiers}},
  note         = {Commit abcdefg • accessed DD Mon YYYY},
  year         = {2025}
}

Release files for verifiers 0.1.15.dev178

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for verifiers 0.1.15.dev178
File	Size	Uploaded
verifiers-0.1.15.dev178.tar.gz	747.7 kB	Jun 11, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for verifiers 0.1.15.dev178
File	Interpreter	ABI	Platform	Reset
verifiers-0.1.15.dev178-py3-none-any.whl 666.6 kB Jun 11, 2026	Python 3	none	any	Details

Total release size: 1.4 MB

Release files / verifiers-0.1.15.dev178.tar.gz

Download URL	verifiers-0.1.15.dev178.tar.gz
Size	747.7 kB
Tags	Source
SHA-256 checksum How to use checksums	`b54747141acd8c62d0ccc26b08a727f7d1b8f206fe70ddaeb8a4d994bfe70f8d`
BLAKE2b-256 checksum How to use checksums	`0a02a44b3c610355f668965459b14fd0f813fe975e073be90d788b071389d1bd`
Upload date	Jun 11, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.13`

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/PrimeIntellect-ai/verifiers
Publishing commit: github.com/PrimeIntellect-ai/verifiers/tree/d13465c0f2d1aebebb6bb154d87fe6f39a50e1a1

Workflow

Publishing configuration: github.com/PrimeIntellect-ai/verifiers/blob/d13465c0f2d1aebebb6bb154d87fe6f39a50e1a1/.github/workflows/publish-verifiers.yml
Publishing logs: github.com/PrimeIntellect-ai/verifiers/actions/runs/27341501301/attempts/1

Artifact

Subject: verifiers-0.1.15.dev178.tar.gz
SHA-256 checksum: b54747141acd8c62d0ccc26b08a727f7d1b8f206fe70ddaeb8a4d994bfe70f8d Verifying attestations

Release files / verifiers-0.1.15.dev178-py3-none-any.whl

Download URL	verifiers-0.1.15.dev178-py3-none-any.whl
Size	666.6 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`893db6d1d07fb597ec3dd59708174dc2afdb17fd449f6a62d807982ca18cdbd5`
BLAKE2b-256 checksum How to use checksums	`c5c6312100f181cd679717bb95b47e95b306bb5a7b8b4759251601c4d17484af`
Upload date	Jun 11, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	Yes
Uploaded via	`twine/6.1.0 CPython/3.13.13`

Provenance

Identity

Publishing platform: GitHub Actions
Publishing repository: github.com/PrimeIntellect-ai/verifiers
Publishing commit: github.com/PrimeIntellect-ai/verifiers/tree/d13465c0f2d1aebebb6bb154d87fe6f39a50e1a1

Workflow

Publishing configuration: github.com/PrimeIntellect-ai/verifiers/blob/d13465c0f2d1aebebb6bb154d87fe6f39a50e1a1/.github/workflows/publish-verifiers.yml
Publishing logs: github.com/PrimeIntellect-ai/verifiers/actions/runs/27341501301/attempts/1

Artifact

Subject: verifiers-0.1.15.dev178-py3-none-any.whl
SHA-256 checksum: 893db6d1d07fb597ec3dd59708174dc2afdb17fd449f6a62d807982ca18cdbd5 Verifying attestations