Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.
Python agentaichatgptdeepseekfine-tuninggemmaimage-generationllamallmllmsopenaipythonqwenreinforcement-learningself-hostedstable-diffusiontext-to-speechttsuiunsloth
Commits on main
torchcodec: cover torch 2.11 in the compatibility guard and pin per torch minor (#7474)
da4f0b9 Daniel Han committed 2 months ago
Studio: keep the uploaded eval dataset for Whisper and audio VLM runs too (#10482)
2495911 Daniel Han committed 2 months ago
Studio: fix sending an image with no caption on Anthropic (#10455)
88888e7 Nilay committed 2 months ago
Studio: keep an Ollama Modelfile SYSTEM prompt in force when the date line is on (#10463)
db2d0ad Hugh Scott committed 2 months ago
Stop unsloth train from ignoring config keys it does not know (#10452)
7b69166 Nilay committed 2 months ago
Make the venv-hardening decode strict, and say which codepoints differed (#10432)
6dcfd43 Daniel Han committed 2 months ago
Pin tokenizers with transformers so Apple Silicon keeps Train and Export (#10431)
9735ec0 Daniel Han committed 2 months ago
Studio: stop re-fetching every llama.cpp release from the GitHub API (#10461)
6ef5955 Nilay committed 2 months ago
Drop late inference responses after their request is released (#10388)
1112620 Ian committed 2 months ago
Docker Studio: build the labextension with @jupyter/builder so it never reaches for GitHub (#10483)
2cd91fa Daniel Han committed 2 months ago
notebook_validator: read shell the way bash does, and state the torch/torchcodec contract (#10414)
6a0966a Daniel Han committed 2 months ago
Studio: name the llama.cpp backend in the install log, and stop labelling Windows ROCm torch as CPU (#10370)
9b30276 Daniel Han committed 2 months ago
Decode PowerShell stdout as UTF-8 in the Windows venv hardening tests (#10476)
9fa8d60 Daniel Han committed 2 months ago
Studio: let Deep Research write longer reports on saved connections (#10254)
ec0cb88 Nilay committed 2 months ago
Studio: add GPT-6 Astra to the fallback catalog and reasoning controls (#10475)
06263ed Michael Han committed 2 months ago
Stop unsloth start codex and openclaw from inheriting provider API keys (#10316)
05d98ea Nilay committed 2 months ago
OpenAI stream: gate UI control frames behind X-Unsloth-Events opt-in (#10362)
f47547b Denys Bushuliak committed 2 months ago
Let unsloth.models and unsloth.save import on Apple Silicon (#9447)
97d67a3 Vineeth Sai Varikuntla committed 2 months ago
Studio: forward the reasoning_effort levels a local template advertises (#10458)
5c44c8c Nilay committed 2 months ago
studio: use the hermes and deepseek logos in the agents picker (#10266)
66178d5 Maheswar Kumar committed 2 months ago
Studio: use the eval dataset for TTS and codec audio training (#10259)
41d2ff4 Nilay committed 2 months ago
Studio: convert pasted images when tools are turned on (#10094)
502c4d2 Nilay committed 2 months ago
Studio: admit a bearer that carries no token under keyless API access (#10423)
acad3f0 Long Yixing committed 2 months ago
Studio: dial the bound address for in-process API calls, not a loopback literal (#10140)
6133092 Long Yixing committed 2 months ago
Studio: stop rejecting provider API keys longer than 190 bytes (#10421)
e0f8415 Long Yixing committed 2 months ago