{"full_name":"unslothai/unsloth","owner":"unslothai","name":"unsloth","url":"https://github.com/unslothai/unsloth","description":"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.","homepage":"https://unsloth.ai/docs","topics":[],"language":"Python","license":"Apache-2.0","avatar":"https://avatars.githubusercontent.com/u/150920049?v=4","created_at":"2023-11-29T16:50:09Z","pushed_at":"2026-11-10T09:38:48Z","metrics":{"stars":79275,"forks":7382,"watchers":380,"issues":5184,"prs":11603,"releases":86,"issues_open":null,"issues_closed":null,"prs_open":null,"prs_closed":null,"prs_merged":null,"discussions":232},"growth":{"day":{"stars":47,"forks":4,"base_day":"2026-11-09"},"week":{"stars":312,"forks":28,"base_day":"2026-11-03"},"month":{"stars":2044,"forks":268,"base_day":"2026-10-11"}},"history":{"dates":["2026-09-19","2026-09-20","2026-09-21","2026-09-22","2026-09-23","2026-09-24","2026-09-25","2026-09-26","2026-09-27","2026-09-28","2026-09-29","2026-09-30","2026-10-01","2026-10-02","2026-10-03","2026-10-04","2026-10-05","2026-10-06","2026-10-07","2026-10-08","2026-10-09","2026-10-10","2026-10-11","2026-10-12","2026-10-13","2026-10-14","2026-10-15","2026-10-16","2026-10-17","2026-10-18","2026-10-19","2026-10-20","2026-10-21","2026-10-22","2026-10-23","2026-10-24","2026-10-25","2026-10-26","2026-10-27","2026-10-28","2026-10-29","2026-10-30","2026-10-31","2026-11-01","2026-11-02","2026-11-03","2026-11-04","2026-11-09","2026-11-10"],"stars":[75854,75946,75994,76028,76065,76158,76205,76239,76308,76377,76439,76537,76585,76639,76710,76773,76828,76929,77026,77119,77133,77169,77231,77271,77343,77506,77626,77707,77876,77940,78002,78064,78121,78184,78269,78322,78375,78425,78474,78521,78595,78631,78694,78756,78819,78963,79008,79228,79275],"forks":[6901,6910,6921,6928,6932,6941,6948,6946,6955,6970,6980,6996,7007,7016,7025,7041,7052,7070,7083,7093,7090,7097,7114,7119,7129,7148,7160,7173,7194,7197,7203,7209,7216,7224,7241,7245,7255,7263,7268,7283,7295,7301,7306,7316,7322,7354,7360,7378,7382],"watchers":[382,382,382,382,384,384,384,384,385,384,384,384,384,385,385,385,385,385,385,384,384,384,385,387,387,388,389,389,386,385,385,387,387,386,387,386,386,386,386,381,382,382,381,380,378,379,379,380,380],"issues_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null],"prs_open":[null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null,null]},"boards":{},"same_owner":[{"full_name":"unslothai/notebooks","name":"notebooks","boards":{}}],"issues":[{"number":2755,"title":"[FEATURE] TPU Support for Fine Tuning","comments":13,"created_at":"2025-06-18T00:53:47Z"},{"number":16317,"title":"[Bug] Unsloth North Mini Code giving Stream interrupted error when using Bearer Token and OpenAIAPI with OpenCode","comments":6,"created_at":"2026-10-31T17:20:46Z"},{"number":9037,"title":"[Bug] Unsloth Studio / Desktop: long chats redo the full prefill after a model reload because slot KV is only saved on idle auto-unload","comments":6,"created_at":"2026-08-16T19:04:05Z"},{"number":5764,"title":"[Feature Request] Multi GPU fine tuning via Unsloth Studio","comments":6,"created_at":"2026-05-25T10:11:22Z"},{"number":4380,"title":"[Feature] Flatpak support - Unsloth Studio","comments":6,"created_at":"2026-03-17T20:20:13Z"},{"number":2739,"title":"[Feature] Add ABBA: Highly Expressive Hadamard Product Adaptation for Large Language Models","comments":6,"created_at":"2025-06-12T14:30:20Z"},{"number":14487,"title":"[Bug] potential memory leak Drive C loses 200-400 mb after every use of unsloth with no way to clear residual data","comments":5,"created_at":"2026-10-21T10:29:21Z"},{"number":10345,"title":"Human-centric context management","comments":5,"created_at":"2026-09-05T09:47:33Z"}],"discussions":[{"number":16817,"title":"Unsloth Desktop issues with using Qwen 3.8 27B with NVFP4","comments":2,"category":"Q&A","created_at":"2026-11-05T15:32:42Z"},{"number":16779,"title":"OpenUI support","comments":0,"category":"Ideas","created_at":"2026-11-05T11:50:08Z"},{"number":16487,"title":"When will Unsloth exit Beta state?","comments":0,"category":"Q&A","created_at":"2026-11-02T20:01:44Z"},{"number":16460,"title":"Run big llm-models like Qwen3.8-Flash-Next in q3 or iq4 like on strata/poystrata on 48GB+16GBVRam","comments":0,"category":"Q&A","created_at":"2026-11-02T06:54:05Z"},{"number":16456,"title":"load error","comments":1,"category":"Q&A","created_at":"2026-11-02T05:21:25Z"},{"number":15852,"title":"VRAM & KV-Cache Estimator for GGUF Export & Post-Fine-Tuning Serving (Zero-Dependency CLI & Interactive WebApp)","comments":0,"category":"Show and tell","created_at":"2026-10-28T20:15:02Z"},{"number":15404,"title":"Is eval_loss computed after response-only masking? (train_on_completions=true, alpaca template)","comments":6,"category":"Q&A","created_at":"2026-10-26T17:17:09Z"},{"number":15102,"title":"use lora in imagegen, textgen, audiogen, videogen","comments":0,"category":"Ideas","created_at":"2026-10-24T08:36:31Z"}],"latest_release":{"name":"Large Performance Fixes","tag":"v0.1.916-beta","published_at":"2026-11-06T14:43:33Z","body":"This release brings large performance fixes across Unsloth: faster typing and search in long chats, faster image/video loads, two-GPU image/video gen. It also adds separate browsers for each chat, floating browser window.\r\n\r\n## Highlights\r\n\r\n- **Large performance fixes** with more Glass theme optimizations coming next week\r\n- **Glass theme fixes** no more blur, better rendering, more options\r\n- **Unified memory offloading** now properly works for big models\r\n- **Separate browsers for each chat** - Multiple chats can now run their own browser individually\r\n- **Floating browser** you can move and resize over the chat & let the model do it's thing\r\n- **Dynamic V3.1 Qwen3.8-Flash** - 50% less token use, 15% more accurate\r\n- **Two-GPU image and video** up to 2x faster on matching NVIDIA cards\r\n- **Faster chat search and deleting** - search and delete chats much faster\r\n- **Stable diffusion image/video models loads** up to 40% faster\r\n- **LTX-2.5 video** much faster for new prompts on 16 to 24 GB GPUs\r\n\r\n## Chat performance\r\n\r\n- Typing in a long chat no longer lags when a Glass theme or chat background is on.\r\n- Chat search stays fast on very long histories instead of failing to load.\r\n- Clearing and deleting chats is much faster on large histories.\r\n- The first use of voice or audio features after launch no longer freezes a streaming reply for seconds.\r\n- Read aloud with the **Unsloth model** engine no longer waits seconds on a model lookup before speaking.\r\n- Glass themes over a background image respond up to 2x faster without GPU rendering, as on some Linux desktops.\r\n\r\n## Images and video\r\n\r\n- With two matching, idle NVIDIA GPUs, models such as SDXL, Qwen-Image and Wan2.2 5B run half of each step on the second card when they fit on one, about 1.5 to 1.8x faster.\r\n- On the same setup, FLUX and Lumina 2 split an even-sized batch of images across both cards.\r\n- SDXL loads up to 40% faster, so the first image arrives sooner.\r\n- On 16 to 24 GB GPUs, LTX-2.5 encodes new prompts on the GPU instead of the CPU, so a clip with a new prompt renders much faster.\r\n- LTX-2.5 GGUF Q8_0 video renders about 10% faster on NVIDIA GPUs, with identical frames.\r\n- Repeat renders with the same LoRA are faster when the model is not quantized, GGUF or offloaded.\r\n- Some image models, such as FLUX and Z-Image, render faster on Linux with AMD RDNA3 and RDNA4 GPUs.\r\n\r\n## New in Unsloth\r\n\r\n- Float the built-in browser over the chat as a small window you can move and resize; each chat remembers whether it floats.\r\n- Load Anima, Chroma, MOSS-Image and Krea 2 image models from single-file checkpoints, with LoRA and LoKR files for all but Chroma.\r\n- Glass themes dim the background image instead of blurring it, and controls on other pages match the chat's glass.\r\n- Owners can rename a managed account without losing its chats, files or settings.\r\n- Owners can cap model download speed in **Settings > General > Downloads**.\r\n- OpenAI-compatible connections can send custom headers.\r\n\r\n## Fixes\r\n\r\n- On chat-only installs with a non-GGUF model already downloaded, the first chat fetches the default model instead of failing.\r\n- GGUF models on CPU-only machines no longer run out of memory with Auto context.\r\n- On Windows, HTTPS downloads resume after a pause or interruption instead of starting over.\r\n- Updating Unsloth on Windows no longer sits silent for half a minute at the end.\r\n- Connecting Blender through MCP works again.\r\n- Compare mode no longer leaves one pane empty on the first message.\r\n\r\n## Training and library\r\n\r\n- In Python, `model.generate()` for Llama, Qwen2 and Qwen3 models is faster by default on NVIDIA GPUs under Linux, up to several times for small models.\r\n- Training on a Hub dataset split into hundreds of files starts up to a minute sooner.\r\n- With **Assistant completions only**, answers that start with a newline are no longer left out of training.\r\n\r\n## Download Unsloth Desktop\r\n\r\nUnsloth Desktop is **free and open source**. Download it for:\r\n