torchtune
PyTorch-native fine-tuning library from Meta, no longer actively maintained (development wound down in 2025)
torchtune is a PyTorch-native fine-tuning library by Meta. It supports full fine-tuning, LoRA, and QLoRA with memory-efficient training that works on consumer GPUs (24GB VRAM). Covers Llama, Mistral, Gemma, Phi, and Qwen model families. Includes recipes for SFT, DPO, and knowledge distillation with built-in evaluation.
Pricing: Free
Choosing a fine-tuning platform right now?
One click, anonymous. Helps us know our visitors.
Thanks, that helps.
Choosing a fine-tuning platform right now?
One click, anonymous. Helps us know our visitors.
Thanks, that helps.
torchtune Alternatives
Explore 20 products in the Fine-tuning category. View all torchtune alternatives.
LLaMA-Factory
Open-source fine-tuning framework for 100+ LLMs with a web UI
Axolotl
Open-source toolkit for fine-tuning LLMs with a single YAML config across the full training pipeline
TRL
Hugging Face library for training language models with RLHF, SFT, and DPO
OVHcloud AI
European cloud provider with AI inference, training, and deployment services
Amazon Bedrock
Managed API access to foundation models on AWS with built-in fine-tuning and agent tooling
Work on torchtune? Feature it at the top of Fine-tuning.
Is your product missing?