Post Training Python Packages

vsa

A unified inference and post-training framework for accelerated video generation.

st-attn

A unified inference and post-training framework for accelerated video generation.

whileai

Scientific RL and SFT post-training for AI agents: build evals that can fail, simulate situations, judge checked against people, pass@1 with a 95% interval, train with GRPO/SFT/DPO, prove every gain on a held-out set. uv add whileai

zeroproof

Scientific RL and SFT post-training for AI agents: build evals that can fail, simulate situations, judge checked against people, pass@1 with a 95% interval, train with GRPO/SFT/DPO, prove every gain on a held-out set. uv add whileai

freesolo-flash-dev

LoRA post-training for open-weight models: SFT, GRPO, and on-policy distillation. Describe a run in TOML; Flash allocates a GPU, trains, streams checkpoints, and serves the adapter.

fastvideo-kernel

A unified inference and post-training framework for accelerated video generation.

freesolo-flash

LoRA post-training for open-weight models: SFT, GRPO, and on-policy distillation. Describe a run in TOML; Flash allocates a GPU, trains, streams checkpoints, and serves the adapter.

rapidfireai

RapidFire AI: Rapid AI Customization from RAG to Fine-Tuning

miniverl

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

fastvideo

A unified inference and post-training framework for accelerated video generation.

telescope-ui

Real-time observability dashboard for the Telescope RL post-training framework. Monitor metrics, rollouts, traces, GPU infrastructure, and evals at scale.

arctic-platform

Arctic Training and Inference Platform

probe-the-harness

Setup checks for stale-data RL comparisons in language models

rldoctor

Diagnose why your GRPO/RLVR run is burning GPU-hours without learning. Reads the log you already have: W&B, TensorBoard, verl, TRL trainer_state.json.

areno

An easy-to-use, fast toolkit to scale up RL post-training on a single node.

autosynth

Agentic synthetic-data generation framework inspired by Meta FAIR's Autodata / Agentic Self-Instruct.

crucible-rl

Turn any real software into a replayable RL environment for training AI agents — deterministic replay, verifiable rewards, TRL & verifiers adapters.

oxrl

A lightweight post-training framework for LLMs and VLMs. 51 algorithms, 38 verified models. Scales with DeepSpeed, vLLM, and Ray.

zeroproof-simulations

Scientific RL and SFT post-training for AI agents: build evals that can fail, simulate situations, judge checked against people, pass@1 with a 95% interval, train with GRPO/SFT/DPO, prove every gain on a held-out set. uv add whileai

polgrad

Reference semantics, conformance testing, and pathology diagnostics for LLM policy-gradient post-training

shadow-peft

Parameter-Efficient Fine-Tuning For Edge-Cloud Collabrative Computing

prefscope

An open-source framework for analyzing behavioral patterns in post-training datasets and language-model responses using sparse autoencoders and model-based interpretation.