Tools to Fine-Tune LLMs Locally

Fine-tune a language model with LoRA, QLoRA or full training on your GPUs, using Unsloth, LLaMA-Factory or TRL.

68 tools
Favicon of NeMo RL

NeMo RL

2 videos
An open-source model post-training library for your own GPUs, with reinforcement learning, fine-tuning and PyTorch or Megatron Core backends.

2.1KUpdated 2 hours agoApache-2.0

Docker#Distributed execution#Hugging Face integration#LoRA

An open-source LLM training platform that prepares data, evaluates and deploys models on your infrastructure, with a hosted service for automated development.

9.4KUpdated 1 day agoApache-2.0

Docker#Distributed execution#LoRA#Multimodal input

An open-source diffusion model training suite supports image and video fine-tuning, with a self-hosted web interface and an MIT license.

12.7KUpdated 8 hours agoMIT

macOS · Windows · Linux · Web#Hugging Face integration#Image-to-image#LoRA

Favicon of TRL

TRL

8 videos
An open-source LLM fine-tuning library trains Transformers models on your GPUs, with preference training, LoRA and Apache 2.0 licensing.

19.8KUpdated 4 hours agoApache-2.0

#Distributed execution#LoRA#Quantization

Favicon of npcpy

npcpy

1 video
An open-source Python AI agent library supports local LLMs and cloud APIs, with multi-agent teams, knowledge graphs and structured output.

1.5KUpdated 1 day agoMIT

macOS · Windows · Linux#Code execution#Knowledge graphs#llama.cpp backend

A local LLM desktop app runs GGUF models offline on Windows, macOS and Linux, with vision chat and OpenAI- and Anthropic-compatible APIs.

47.7KUpdated 3 months agoAGPL-3.0

macOS · Windows · Linux · Docker · Web#GGUF#llama.cpp backend#LoRA

A self-hosted Kubernetes operator deploys LLMs with vLLM, provisions GPU resources, and manages fine-tuning and document retrieval services.

1KUpdated 4 hours ago

#Distributed execution#Hugging Face integration#LoRA

Favicon of Kubeflow

Kubeflow

1 video
A self-hosted AI platform for Kubernetes that combines model training, portable ML workflows and model management under Apache 2.0.

16KUpdated 2 months agoApache-2.0

Web#MLX#Multi-user access

Self-hosted LLM platform with OpenAI-compatible APIs, a browser chat UI and model evaluation tools. Open source under Apache 2.0.

39.5KUpdated 1 year agoApache-2.0

macOS · Linux · Web#Batch processing#Distributed execution#Hugging Face integration

Local AI fine-tuning library for Apple Silicon Macs, built on MLX with an Unsloth-compatible API. Open source under Apache 2.0.

1.5KUpdated 5 months agoApache-2.0

macOS#GGUF#Hugging Face integration#LoRA

A local AI tool trains a personal model on your memories, with roleplay and permission-based connections to other AI identities and apps.

15.7KUpdated 1 year agoApache-2.0

macOS · Windows · Linux · Docker · Web#llama.cpp backend#MLX#Multi-agent workflows

An open-source local LLM framework runs and fine-tunes large MoE models across CPUs and GPUs, with SGLang and LlamaFactory integrations.

19.7KUpdated 6 days agoApache-2.0

Docker#LoRA#Multimodal input#Prompt caching

Favicon of AgileRL

AgileRL

1 video
Open-source Python framework for reinforcement learning and LLM fine-tuning. Train on your own GPU, with Apache 2.0 licensing and multi-GPU support.

971Updated 12 hours agoApache-2.0

Linux#Distributed execution#Hugging Face integration#LoRA

An open-source PyTorch library fine-tunes LLMs on your own GPUs with LoRA and QLoRA support, though development is no longer active.

5.8KUpdated 1 week agoBSD-3-Clause

#Hugging Face integration#LoRA#Quantization

Favicon of StarCoder2

StarCoder2

1 video
Local code completion models run on CPUs or GPUs, work with Hugging Face Transformers, and support LoRA fine-tuning.

2.1KUpdated 3 years agoApache-2.0

#Hugging Face integration#LoRA#Quantization

An open-source financial LLM framework analyzes news and forecasts stock movements, with local models or cloud inference through OpenAI and MiniMax.

21.6KUpdated 14 hours agoMIT

Web#Hugging Face integration#LoRA#Multilingual

MLX VLM

8 videos
A local AI package runs and fine-tunes multimodal models on Macs using MLX, with an OpenAI-compatible server and support for image, audio and video tasks.

5.9KUpdated 9 hours agoMIT

macOS · Web#Agent Skills#Batch processing#Hugging Face integration

A local conversational LLM that supports tool requests, CPU and NVIDIA GPU inference, and quantized models for lower memory use.

12.4KUpdated 2 months agoApache-2.0

Linux · Web#Distributed execution#Hugging Face integration#Multilingual

Favicon of Needle

Needle

5 videos
An on-device AI model turns text into tool calls, extracts structured data and creates local search embeddings on phones, wearables and computers.

14.8KUpdated 11 hours agoApache-2.0

macOS · Windows · Linux · iOS · Android · Web#Hugging Face integration#LoRA#Multilingual

Open-source reinforcement learning infrastructure for LLMs and agents, with asynchronous training, PyTorch and Megatron backends, and Apache 2.0 licensing.

5.9KUpdated 17 hours agoApache-2.0

#Distributed execution#Hugging Face integration#LoRA

An open-source AI training and inference framework for your own GPUs, with distributed parallelism, memory management and Apache 2.0 licensing.

41.4KUpdated 12 hours agoApache-2.0

#Distributed execution

A self-hostable MLOps platform that tracks experiments, versions datasets and automates ML jobs, with model serving backed by NVIDIA Triton.

6.9KUpdated 2 weeks agoApache-2.0

Docker · Web#Code execution#Git integration#Multi-user access

Favicon of RD Agent

RD Agent

1 video
A self-hosted AI research agent for Linux that writes and tests model experiments. MIT licensed, with chat and embedding calls through LiteLLM providers.

15.4KUpdated 2 months agoMIT

Linux · Web#Code execution

An open source LLM training library that runs on your own hardware, supports the Tinker API, and includes environments for training tool-using agents.

2.5KUpdated 9 hours agoApache-2.0

Docker#Code execution#Distributed execution#Tool calling

Favicon of Unsloth

Unsloth

23 videos
A local LLM app that runs and trains models on macOS, Windows and Linux, with media generation and connections to coding agents.

79.7KUpdated 1 day agoApache-2.0

macOS · Windows · Linux · Docker · Web#Code execution#GGUF#Image-to-image

Favicon of DSPy

DSPy

6 videos
A Python framework builds and optimizes local LLM applications with Ollama on laptops or SGLang on GPU servers. Open source under the MIT license.

39.1KUpdated 13 hours agoMIT

#Code execution#MCP#Multimodal input

Favicon of Soup

Soup

2 videos
Local LLM fine-tuning software with a browser dashboard, Hugging Face model support, and training on CUDA or Apple Silicon GPUs.

10.6KUpdated 4 hours agoApache-2.0

macOS · Windows · Linux · Docker · Web#Batch processing#Distributed execution#GGUF

Favicon of LitGPT

LitGPT

1 video
An open-source LLM toolkit trains, fine-tunes and serves models on your own hardware, with quantization and multi-GPU support.

13.8KUpdated 1 week agoApache-2.0

#Hugging Face integration#LoRA#Quantization

A local vision-language model connects text with specific image regions, uses Vicuna checkpoints, and runs GPU inference with a Gradio browser interface.

8.6KUpdated 2 months ago

Web#Multimodal input

Favicon of PEFT

PEFT

4 videos
A Python library for fine-tuning LLMs and diffusion models with less compute and storage, using adapters and quantization under Apache 2.0.

21.9KUpdated 12 hours agoApache-2.0

macOS#Distributed execution#Hugging Face integration#LoRA

An open-source AI agent training framework using reinforcement learning with existing agent harnesses, local execution and Kubernetes support.

19KUpdated 1 week agoMIT

#Code execution#MCP#Multimodal input

Favicon of Nemotron

Nemotron

29 videos
A family of AI models for self-hosted agents, with open weights and training data, supporting Ollama and llama.cpp on NVIDIA GPUs.

2.4KUpdated 2 weeks agoApache-2.0

Linux#GGUF#Guardrails#Hugging Face integration

Favicon of MLX LM

MLX LM

10 videos
A local LLM Python package runs text generation and fine-tuning on Apple Silicon with MLX, Hugging Face model support and quantization.

7.7KUpdated 2 days agoMIT

macOS#Batch processing#Distributed execution#Hugging Face integration

An open-source LLM training library for GPU workstations and clusters, with Hugging Face compatibility, LoRA fine-tuning and multimodal training.

1.1KUpdated 8 hours agoApache-2.0

Docker#Distributed execution#Hugging Face integration#LoRA

Favicon of LogotyperLogotyper
Got an idea for an AI tool? I'm Eugene, the designer and developer behind Localhosted. I'll help you build it, from the first sketch to launch.
Start your project
Favicon of Logotyper
Favicon of LogotyperLogotyper
Got an idea for an AI tool? I'm Eugene, the designer and developer behind Localhosted. I'll help you build it, from the first sketch to launch.
Start your project
Favicon of Logotyper

More in Fine-Tuning and Training