# fine-tuning

ModelScope's fine-tuning framework supporting 600+ models

ms-swift is ModelScope's open-source framework for fine-tuning over 600 large language and multimodal models. It supports SFT, DPO, RLHF, LoRA, QLoRA, and full fine-tuning with a web UI and CLI interface. Optimized for the Chinese AI ecosystem with native ModelScope Hub integration alongside Hugging Face support. Over 13,500 GitHub stars.

Meta's official PyTorch library for LLM fine-tuning

torchtune is Meta's official PyTorch-native library for fine-tuning large language models. It provides composable building blocks for training recipes covering LoRA, QLoRA, full fine-tuning, DPO, and knowledge distillation. Supports Llama, Mistral, Gemma, Qwen, and Phi model families with distributed training across multiple GPUs. Designed as a hackable, dependency-minimal alternative to higher-level frameworks.

Unified framework for fine-tuning 100+ large language models

LLaMA-Factory is an open-source toolkit providing a unified interface for fine-tuning over 100 LLMs and vision-language models. It supports SFT, RLHF with PPO and DPO, LoRA and QLoRA for memory-efficient training, and continuous pre-training. The LLaMA Board web UI enables no-code configuration, while CLI and YAML workflows serve advanced users. Integrates with Hugging Face, ModelScope, vLLM, and SGLang for model deployment.

2x faster LLM fine-tuning with 70% less VRAM on a single GPU

Unsloth is an open-source framework for fine-tuning large language models up to 2x faster while using 70% less VRAM. Built with custom Triton kernels, it supports 500+ model architectures including Llama 4, Qwen 3, and DeepSeek on consumer NVIDIA GPUs. Unsloth Studio adds a no-code web UI for dataset creation, training observability, model comparison, and GGUF export for Ollama and vLLM deployment.

API for GPT-4, o1, DALL-E, Whisper, and embeddings

Official API platform for GPT-4o, o1/o3 reasoning models, DALL-E image generation, Whisper speech-to-text, and text embeddings. Features Assistants API, function calling, JSON mode, fine-tuning, and batch processing. The most widely used AI API in the industry, powering millions of applications from chatbots to complex multi-step agent systems across every sector.

Google Cloud ML platform with Gemini and custom models

Google Cloud's end-to-end ML platform with Gemini models, Model Garden featuring 150+ models, AutoML, and custom training pipelines. Features Vertex AI Search, Conversation, and Agent Builder for enterprise AI applications. The comprehensive platform for organizations building production AI systems at scale within the Google Cloud ecosystem, with enterprise governance and compliance built in.

The GitHub of ML — model hub, datasets, and inference

Open-source platform for building, sharing, and deploying machine learning models and datasets. Hosts 500k+ models, 100k+ datasets, and Spaces for interactive demos. The central hub of the open-source AI ecosystem, providing model discovery, inference APIs, and collaborative tools that make it the GitHub of machine learning for researchers and developers worldwide.

Run and deploy ML models via API with simple pricing

Unified API gateway that provides access to hundreds of LLM models from OpenAI, Anthropic, Google, Meta, and open-source providers through a single OpenAI-compatible interface. Features model fallbacks, price comparison, and community-driven model rankings. The most popular LLM routing service for developers who want multi-provider flexibility without managing individual API integrations.

Production-grade inference with serverless and on-demand GPUs

Open-source model serving platform optimized for large language models and generative AI. Supports Hugging Face models, LoRA adapters, and continuous batching for efficient multi-user serving. Built on PyTorch with OpenAI-compatible endpoints. Designed for teams who need production-grade LLM serving with lower latency and better resource utilization than generic model serving frameworks.

Fast inference platform for open-source models

Meta's open-source large language model family available for commercial use. Llama 3 models range from 8B to 405B parameters, offering competitive performance with full weight access. Hosted on Hugging Face and available through major cloud providers. The most impactful open-source AI release, enabling companies and researchers to build, fine-tune, and deploy custom AI solutions without API dependencies.