unslothai/unsloth Alternatives

For developers who want a scriptable Python/Lightning training and inference library without a GUI.

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

Choose it when you want an all-in-one Python/CLI platform for fine-tuning, evaluation, and deployment rather than a desktop app.

Easily fine-tune, evaluate and deploy Qwen, Gemma, or any open weight LLM!

Choose for distributed PyTorch-native training of LLMs/VLMs on multi-GPU clusters, not for local desktop use.

🚀 Pytorch Distributed native training library for LLMs/VLMs with OOTB Hugging Face support

Choose for web/YAML-based PEFT and full-parameter training of 100+ models if you need broad coverage and don't require Unsloth's optimized kernels.

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

Choose for extremely broad LLM/MLLM support (600+ models) with a Python/CLI training workflow.

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.6, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM4.5v, Gemma4, Llava, Phi4, ...) (AAAI 2025).

Choose for a specialized engine aimed at training very large MoE models, not a consumer GUI.

A Next-Generation Training Engine Built for Ultra-Large MoE Models

Choose for very low-memory fine-tuning via layer streaming when your GPU is smaller than Unsloth's typical requirement.

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

Choose if you prefer a simple Python library/CLI for building and personalizing LLMs without a desktop app.

Build, personalize and control your own LLMs. From data pre-processing to fine-tuning, xTuring provides an easy way to personalize open-source LLMs. Join our discord community: https://discord.gg/TgHXuSJEk6

Choose for large-scale JAX/TPU training instead of a local GUI/model runner.

A simple, performant, and scalable Jax LLM!

Choose for a web-based research environment to train/evaluate/scale models across hardware, if you prefer that over a native desktop app.

The open source research environment for AI researchers to seamlessly train, evaluate, and scale models from local hardware to GPU clusters.

Choose for a self-hosted web UI with broad llama.cpp/HF support, vision, tool-calling, and LoRA training instead of a native desktop app.

Open-source desktop app for local LLMs. Text, vision, tool-calling, OpenAI/Anthropic-compatible API. 100% private.

Choose when you only need to run open-weight models locally via CLI/API and do not need fine-tuning or diffusion support.

Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.

Choose for a ready-made offline agent/chat app for local LLMs if inference-only and private deployment are enough.

Local AI app and inference engine for agents. Run open-weight LLMs locally — private, 100% offline on your computer. Join our Discord: https://discord.com/invite/8wGSsvmg4V

Choose for a polished offline ChatGPT-style chat client when you have no need to train or run diffusion models.

Jan is an open source alternative to ChatGPT that runs 100% offline on your computer.

An old, inference-only Llama 2 chat app; choose it only if you specifically need a simple offline Llama-2-era chat.

A self-hosted, offline, ChatGPT-like chatbot. Powered by Llama 2. 100% private, with no data leaving your device. New: Code Llama support!

Choose for fine-tuning Gemma multimodal models on Apple Silicon with PyTorch/MPS, but it is model- and platform-specific.

Fine-tune Gemma 4 and 3n with audio, images and text on Apple Silicon, using PyTorch and Metal Performance Shaders.