Snowgoose provides access to models from OpenAI, Anthropic, Google, and other providers in one AI workspace.
Unsloth
Unsloth Studio provides a local workspace for running and training AI models.
About Unsloth
Unsloth Studio is a locally installed workspace for running, training and exporting open AI models through a no-code interface. It runs GGUF, MLX and diffusion image/video models on Mac, Windows and Linux, and supports training across text, vision, TTS/audio and embedding models using LoRA, FP8, FFT and PT kernels, with the page stating that training runs on NVIDIA hardware. A desktop app is also offered for the same three platforms.
Training datasets can be generated from PDF, CSV, JSON, DOCX, TXT or YAML files rather than requiring a prepared dataset. Models can be exposed as an API endpoint for agent tools such as Claude Code and Codex, exported to GGUF or 16-bit safetensor, and deployed remotely for access over Cloudflare HTTPS. The page labels the product beta and does not state pricing; its speed, accuracy and search-privacy statements are vendor claims, and it does not explain how web search queries are routed.
Key features
- Run GGUF, MLX and diffusion image/video models locally on Mac, Windows and Linux
- Train 500+ text, vision, TTS/audio and embedding models with LoRA, FP8, FFT and PT
- Auto-create training datasets from PDF, CSV, JSON, DOCX, TXT files or YAML configs
- Sandboxed Bash and Python code execution with self-healing tool calling
- Serve local models as an API endpoint for Claude Code and Codex
- Export trained models to GGUF or 16-bit safetensor formats
Best for: Developers and researchers who want to fine-tune and run open models on their own machine without writing training code.
Good for
- Runs and trains models locally on Mac, Windows and Linux, with a separate desktop app
- No-code training with automatic dataset creation from common document formats
- Local models can be connected to external agent tools and deployed for remote access
Watch out for
- The product is described as a beta release
- No pricing information is published on the page
- Speed, accuracy and search-privacy figures are vendor claims without independent verification
Unsloth at a glance
- Website
- unsloth.ai/docs/new/studio
- Type
- Online tool
- Pricing
- Not confirmed. Check the site
- Platforms
- Web
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 6, 2026 · Online when checked
- Listed
- Oct 5, 2026
Unsloth alternatives
Tools and sites you could use instead, ranked by how closely they match.
Kili Technology provides annotation tools for building datasets to train, fine-tune, and evaluate AI models.
Jan provides an open source ChatGPT alternative for running models locally or connecting to cloud models.
Curate and annotate vision, audio, and LLM datasets, track experiments, and manage models on a single platform.
Free trial
24 alternatives
AI/ML API provides access to models for text, images, video, audio, code, and embeddings through one API.
CometAPI provides a unified API and billing interface for accessing hundreds of language, image, and audio models.
Popular in AI Infrastructure & MLOps
Well-known listings from the same topic.
Guild AIguild.aiFlow AIflow-ai.comsystemprompt.iosystemprompt.ioAPI docsdocs.x.aiEyePop.aieyepop.aiWeco Observeweco.ai
Unsloth FAQ
Do I need to prepare a dataset before training?
The page states no dataset is needed and that datasets can be auto-created from PDF, CSV, JSON, DOCX and TXT files, or supplied as YAML configs.
Which operating systems are supported?
Local running of GGUF, MLX and diffusion models is listed for Mac, Windows and Linux, and a desktop app is available for the same platforms.
Can the local models be used with agent tools such as Claude Code or Codex?
Yes. The local models can be connected through Unsloth's API endpoint, and the page also notes that providers such as OpenAI, Anthropic or vLLM can be connected to Unsloth.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.