#3737th of 60 in Training and fine-tuning
transformerlab-app
ML research workbench for training, fine-tuning, evaluating and chatting with models
- Stars
- 5.2k
- License
- AGPL-3.0
- Last commit
- Aug 2026
- Last release
- Jun 2026
- Language
- Python
4,878 on 21 Apr 20265,200 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Transformer Lab is a web UI and Python backend for downloading LLMs, fine-tuning them (LoRA/QLoRA, DPO, ORPO, SIMPO), evaluating them, and running diffusion and TTS models. It runs on a single machine with MLX, CUDA or ROCm trainers. A Teams edition submits jobs to Slurm or SkyPilot clusters from the same UI and CLI.
Who it is for: ML researchers and labs fine-tuning models on local GPUs or clusters
Strengths
- Covers training, evaluation, chat, diffusion and TTS in one UI
- Trainers for Apple Silicon (MLX), NVIDIA (CUDA) and AMD (ROCm)
- Teams edition schedules jobs on Slurm or SkyPilot clouds (AWS, GCP, Azure)
- Lab SDK and CLI add job tracking to existing Python training scripts
Weaknesses
- Requires Apple Silicon, or an NVIDIA or AMD GPU; no CPU-only path listed
- AGPL-3.0 license may restrict use in proprietary or hosted products
- Windows support only through WSL2 and only with NVIDIA GPUs
- Install is a curl-piped shell script into ~/.transformerlab
What it needs
- GPU required
- Compose
- Needs Conda, Node.js 22 (frontend development), Slurm or SkyPilot (Teams edition)
- Models: Llama 3, DeepSeek, Mistral, Qwen, Phi
- port 8338
What changed
- v0.39.0 ↗ (opens in a new tab)Security fix
“Update js-cookie to fix vulnerability by @dadmobile in https://github.com/transformerlab/transformerlab-app/pull/2203”
From the release notes. No flag means the note does not say so, not that the upgrade is safe.
Also in Training and fine-tuning
See all 60| Rank | Project | Score |
|---|---|---|
| 1 | peftPython library for LoRA and other parameter-efficient fine-tuning of Transformers models | 89 out of 100 |
| 2 | unslothDesktop and web app to run and fine-tune local models | 88 out of 100 |
| 3 | trlPost-training library with SFT, DPO, GRPO and KTO trainers | 87 out of 100 |
| #4 | deepspeedPyTorch library for distributed training of very large models with ZeRO | 86 out of 100 |
| #5 | megatron-lmDistributed training library for large transformer models on NVIDIA GPUs | 85 out of 100 |