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
Studio: keep an API key from borrowing the operator's Hugging Face token (#10126) (#10142)
052b2fa Harsha committed 2 months ago
Kaggle T4: describe the batched-generation expectation as what it is (#10377)
c25841f Daniel Han committed 2 months ago
Docker: one command sets up the NVIDIA Container Toolkit on the host (#10366)
70c336b Daniel Han committed 2 months ago
Kaggle T4: fail a batched row that generates nothing while its single does not (#10367)
96171a1 Daniel Han committed 2 months ago
Studio: keep the overlay rail honest about what it reserves (#10372)
7ba00bb Daniel Han committed 2 months ago
Studio: unsloth chat loads the model you picked, not another GGUF in the same folder (#10357)
76d25ae Nilay committed 2 months ago
studio: stop find in page matching part way through a grapheme (#10200)
a7dec18 Daniel Han committed 2 months ago
deps: bump setuptools build pins and raise the ip-address override to 10.3.1 (#10138)
7a00795 Daniel Han committed 2 months ago
Chat: keep the payload when a provider streams arguments as a decoded object (#10275)
b8bfff2 Shubhankar Tripathy committed 2 months ago
Bump the npm-frontend-security group across 1 directory with 5 updates (#10194)
1e298e0 dependabot[bot] committed 2 months ago
Studio: measure the real size of attached documents instead of guessing (#10318)
2089a25 Nilay committed 2 months ago
Studio: prefer the self-contained MTP head so llama-server's --fit can measure it (#10342)
353aa33 Daniel Han committed 2 months ago
Studio: take GGML_CUDA_ENABLE_UNIFIED_MEMORY only when the weights outgrow the APU carve-out (#10351)
9a21df4 Daniel Han committed 2 months ago
Docker publish: keep the tag list to core, latest, studio, releases and dated nightly pins
cc0cdab Daniel Han committed 2 months ago
Docker: retry the deadsnakes PPA add so a Launchpad 504 does not fail the build
de3a347 Daniel Han committed 2 months ago
Docker: print the Studio admin password in docker logs, add UNSLOTH_STUDIO_PASSWORD (#10344)
6816025 Daniel Han committed 2 months ago
Make the credential probe actually delete its throwaway tag (#10294)
5af6fe1 Daniel Han committed 2 months ago
Revert "Studio: keep a Downloads entry when the list is empty" (#10298)
ecb2083 Etherl committed 2 months ago
Docker publish: sync the Hub page with DOCKER_API_KEY on the namespace route (#10343)
7deb244 Daniel Han committed 2 months ago
Restore facts the comment trim cut from the studiobench harness (#10331)
2acc200 Daniel Han committed 2 months ago
Restore facts the comment trim cut from the Studio backend (#10329)
f2b105e Daniel Han committed 2 months ago
Restore facts the comment trim cut from the unsloth package (#10328)
ca42c01 Daniel Han committed 2 months ago
Stop asserting the titlebar navigation padding from source (#10326)
71d8015 Daniel Han committed 2 months ago
Docker: describe the shipped images on Docker Hub and in the README (#10334)
6a5a644 Daniel Han committed 2 months ago
Docker publish: verify the merged manifest against the arches' children (#10333)
b080a07 Daniel Han committed 2 months ago
Studio: fix whisper prebuilt pairing on linux-arm64 with a GPU llama runtime (#10278)
a4c815d Daniel Han committed 2 months ago
Install the cublas wheel NVIDIA actually publishes for the bundle's CUDA major (#10324)
7197da2 Daniel Han committed 2 months ago
Restore the facts the frontend comment trim reworded away (#10292)
55418be Daniel Han committed 2 months ago
Add Unsloth Docker images (base + Studio) for any NVIDIA GPU host, Ampere through Blackwell (#5748)
5825574 Daniel Han committed 2 months ago