@UnslothAI

Run and train models locally with the Unsloth Desktop app. šŸ¦„ https://xitter.cf/t.co/2kXqhhvdCD

 San Francisco, CA
unsloth.ai
Joined November 2023
  • Tweets806
  • Following484
  • Followers102K
294Media
Introducing Unsloth Desktop šŸ¦„ The first desktop app to run and train models locally. ā€¢ Open-source. Runs on Mac, Windows and Linux ā€¢ Supports MLX, diffusion image/video, audio, GGUF ā€¢ Connect Claude Code and Codex to local LLMs ā€¢ 50% more accurate, self-healing tool calls + sandboxed code exec ā€¢ Works for CPU + multiGPU setups - NVIDIA, AMD, Intel, Mac ā€¢ Train models 2Ɨ faster with 70% less VRAM ā€¢ Private web search, deep research, RAG, MCP and exports (NVFP4, GGUF) ā€¢ Use Unslothā€™s OpenAI-compatible API and cloud models ā€¢ Securely deploy LLMs remotely and access anywhere Unsloth Desktop is now available on unsloth.ai and GitHub. GitHub: github.com/unslothai/unsloth Blog and Guide: unsloth.ai/docs/desktop
286
652
189
4,852
881,631
Run Flint locally on 24GB RAM via Unsloth Dynamic GGUFs! Flint is the strongest MoE model under 50B parameters. Recursiveā€™s Apache-2.0 model has 39.8B parameters (2.5B active). Run at 125 t/s on Mac, 150 on RTX Spark GGUF: huggingface.co/unsloth/flintā€¦ Guide: unsloth.ai/docs/models/flint
Richard Socher
@RichardSocher22h
Introducing Flint, @Recursive_SIā€™s first model: a fast local coding model built with our automated research system, the Eureka Machine. ā€¢ Best-in-class fast local coding ā€¢ Up to 2x faster than comparable models: 125+ tokens/sec on a MacBook Pro, 150+ on RTX Spark ā€¢ Built from scratch in under 4 months Private, unmetered intelligence on your own machine. Open weights available now on @huggingface, run it locally on your favorite framework or try it for free on @opencode or @pidotdev.
37
80
10
740
35,341
Unsloth AI retweeted
Ornith
@ornith_Oct 20
šŸ„³ Ornith-1.5-122B Unsloth GGUFs are now available! šŸ§” Thanks for the awesome work by @UnslothAI šŸ”— huggingface.co/unsloth/Ornitā€¦

unsloth/Ornith-1.5-122B-GGUF Ā· Hugging Face

Weā€™re on a journey to advance and democratize artificial intelligence through open source and open science.

huggingface.co
Ornith
@ornith_Oct 20
Aloha!šŸŒŗ Meet Ornith-1.5-122B: No Tradeoff Between Strong and Fast Stronger. Faster. Leaner. Ornith-1.5-122B delivers frontier-level capabilities with exceptional inference efficiency. With superior instruction following, stronger task decomposition, and deeper understanding of real-world workflows, it achieves state-of-the-art performance among open-source models of comparable size, while running on a single NVIDIA DGX card!! Open weights under the MIT license. Coding āœ…SWE-bench Verified: 0.845 āœ…TB2.1: 0.813 āœ…CyberGym: 0.800 āœ…NL2Repo: 0.512 Agentic āœ…HLE: 0.446 āœ…BrowseComp: 0.785 āœ…MCP-Atlas: 0.746 āœ…Toolathlon: 0.708 šŸ¤—Hugging Face: huggingface.co/ornith-ai/Ornā€¦ We canā€™t wait to see what you build with Ornith-1.5-122B.
5
19
4
176
12,728
DeepSeek V4.1-Flash-1017 can now run locally! šŸ³ We made 1-bit retain ~88.2% accuracy while shrinking size from 508GB to 250GB (āˆ’52%). Run on a 128GB+ RAM device. V4.1 Flash 1017 is the strongest open model to date Guide: unsloth.ai/docs/models/deepsā€¦ GGUF: huggingface.co/unsloth/DeepSā€¦
DeepSeek
@deepseek_aiOct 17
šŸš€ DeepSeek-V4.1-Flash-1017 is here! Stronger benchmark performance, faster task completion, and shorter outputs than the previous versionā€”a Pareto improvement in performance and cost. Now live on our website, app, and API. šŸ› ļø Optimized for @DeepSeekHarness, Tencent CodeBuddy, WorkBuddy, and more. Upgrade to DeepSeek Harness v0.2.1-rc.1 or later for the best experience with the new model. 1/4
123
293
61
3,109
295,516
You can now run and train 120B+ parameter models locally on a laptop! NVIDIA RTX Spark is here. šŸ’š With Unsloth and RTX Spark, train diffusion, decision, multimodal models ~2.7Ɨ faster with optimized kernels and binaries for CUDA on ARM. All local with up to 128GB unified memory
NVIDIA RTX Spark
@NVIDIARTXSparkOct 16
NVIDIA RTX Spark is here. Build AI locally, bring your creative ideas to life, and take your favorite games along. A new beginning for @Windows PCs starts today.
23
34
6
382
26,037
Unsloth AI retweeted
leonie
@helloiamleonieOct 16
how do you fine-tune a decision model on your own data? in these two notebooks, you'll learn: ā€¢ how to prepare the training data ā€¢ how to fine-tune our d1-3B decision model ā€¢ using @UnslothAI and @huggingface TRL all within 30 minutes on a free-tier GPU. unsloth: github.com/Liquid4All/cookboā€¦ TRL: github.com/Liquid4All/cookboā€¦ ps: how adorable are these halloween kitties? šŸ‘»
42
241
14
1,989
83,609
Meta releases Muse Spark Open, a new SOTA open model. Run on a Mac Studio via Dynamic 2-bit Unsloth GGUFs. Muse-Spark-Open-816B-A42B supports text, images, video and audio input with 1M context. GGUF: huggingface.co/unsloth/Muse-ā€¦ Guide: unsloth.ai/docs/models/muse-ā€¦
Mark Zuckerberg
@finkdOct 15
Muse Spark Open -- weights releasing today -- runs great on new Mac Studios thanks to work with @johnternus and Apple.
35
77
16
617
33,734
NVIDIA releases 2 new models: Nemotron-3.5-Bolt and Nemotron-3.5-Super-VL. Nemotron-3.5-Bolt-74B-A9B runs on 24GB RAM. It delivers top-tier performance for its size. Super-VL-120B-A12B runs on 82GB RAM. Bolt GGUF: huggingface.co/unsloth/NVIDIā€¦ Guide: unsloth.ai/docs/models/nemotā€¦ Super GGUF: huggingface.co/unsloth/NVIDIā€¦
NVIDIA AI
@NVIDIAAIOct 15
Meet NVIDIA Nemotron 3.5 Super VL Our new open model combines reasoning and tool use with text, image, document, and video understanding for long-running AI agents. 120B total parameters with 12B active šŸ§µ
48
114
26
1,031
60,366
Video: Decision model trained in ~2 mins on a single DGX Spark using Unsloth! Qwen3.5-2B as base model. Just follow instruction here: unsloth.ai/docs/basics/trainā€¦
12
34
1
326
24,140
You can now train your own Decision model with our free notebook! šŸ’” Qwen3.5-4B will generate decisions instead of text on just 8GB VRAM locally. Learn to data prep (state, questions, gold answers), train, serve. Notebook: colab.research.google.com/giā€¦ Guide: unsloth.ai/docs/basics/trainā€¦
Unsloth AI
@UnslothAIOct 7
You can now train your own Decision model like Jev locally! We increased Qwen3.5 0.8Bā€™s aggregate accuracy from 20.7% to 74.3% across 3 decision benchmarks - on just 4GB VRAM. Turn any LLM like Qwen3.8, Gemma 4 into decision models with our open-source Unsloth repo. We fine-tuned with a Clef head using Unsloth and LoRA (r=64) for one epoch, increasing downstream accuracy from 30ā€“37% to 78%. GitHub: github.com/unslothai/unsloth Guide and Notebooks: unsloth.ai/docs/basics/trainā€¦
108
451
67
3,759
350,351
Windows now has sandboxing! Microsoft released an open-source repo, mxc, for sandboxed code execution. We collaborated with Windows to add mxc OS level sandboxing to Unsloth which adds just <100 ms of overhead. GitHub: github.com/unslothai/unsloth Guide: unsloth.ai/docs/new/studio/sā€¦
40
212
35
1,559
63,460
You can now train your own Decision model like Jev locally! We increased Qwen3.5 0.8Bā€™s aggregate accuracy from 20.7% to 74.3% across 3 decision benchmarks - on just 4GB VRAM. Turn any LLM like Qwen3.8, Gemma 4 into decision models with our open-source Unsloth repo. We fine-tuned with a Clef head using Unsloth and LoRA (r=64) for one epoch, increasing downstream accuracy from 30ā€“37% to 78%. GitHub: github.com/unslothai/unsloth Guide and Notebooks: unsloth.ai/docs/basics/trainā€¦
186
510
84
4,210
335,716
With just 2.5GB VRAM, you can train your own Decision Model using small models like Laya. Training is simply done through a UI interface. Video tutorial and analysis are in our guide. GitHub repo: github.com/unslothai/unsloth

GitHub - unslothai/unsloth: Local UI to run and train LLMs and diffusion models. Supports GGUF,...

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more. - unslothai/unsloth

github.com
5
10
110
9,614
Google releases EmbeddingGemma 2, a new open model that runs locally on 0.5GB RAM. The 740M parameter Apache 2.0 model combines a 270M text model with vision (170M) + audio (300M). Run & train the model via Unsloth. GGUF: huggingface.co/unsloth/embedā€¦ Guide: unsloth.ai/docs/models/embedā€¦
Google DeepMind
@GoogleDeepMindOct 6
Meet EmbeddingGemma 2, our first natively multimodal open model for on-device embeddings. It expands beyond text to unify code, images, audio, and video in a shared space. šŸ§µ
79
401
29
3,555
184,523
Unsloth AI retweeted
Daniel Han
@danielhanchenSep 30
We're co-hosting an open source party with @HuggingFace! šŸ¤—šŸ¦„ Come join us, we'll be handing out @UnslothAI merch, showcasing our upcoming features and more! It'll be a super fun night with lots of demos, DJs, food and more. Join: luma.com/OpenTogether Use code: NOSLOTHSHERE to get in.
7
10
1
115
17,289
You can now run Laya Decision models locally on just 4GB RAM! šŸ”„ Works on CPU, Mac, Windows, Linux and GPU setups. Serve Laya through a Jev-compatible API via Unsloth Desktop. GitHub: github.com/unslothai/unsloth Guide: unsloth.ai/docs/models/decisā€¦
71
317
35
2,426
156,778
We got @UnslothAI a DGX Station! @DanielHanChen and @NaderLikeLadder checked out Unslothā€™s new @Dell Pro Max with GB300 and talked about what comes next: support for more models, faster quantization, and more efficient reinforcement learning.
30
23
9
298
25,605
Unsloth has surpassed 500M model downloads on Hugging Face! šŸ¦„šŸ¤— Qwen3.8-27B GGUF is already Unslothā€™s #1 most-downloaded model ever. Thanks for all your support!
58
67
17
1,070
52,569
Qwen-Image-2.1 can now run locally on 12GB VRAM with Unsloth GGUFs! šŸ–¼ļø The 7B model performs on par with Nano Banana 2.0. For higher quality, you can also run Dynamic FP8 on just 6GB of VRAM via offloading. GGUF: huggingface.co/unsloth/Qwen-ā€¦ Guide: unsloth.ai/docs/models/qwen-ā€¦
Qwen
@Alibaba_QwenSep 20
Meet Qwen-Image-2.1, the most balanced and cost-effective image generation model in the Qwen-Image series! Now open weights! šŸŽØ A unified model for both generation and editing, delivering top-tier quality in a lightweight package. Highlights: šŸ‘€ - Compact & exceptionally fast: A lightweight 7B architecture that outperforms most closed-source models, with drastically accelerated inference for multi-image inputs. - Native transparency: Natively generates and edits RGBA layers, enabling seamless compositing and text editing within transparent images. - Versatile, high-fidelity editing: Supports up to 10 reference images and precise local control while preserving strict fidelity for portraits and products. - Broad coverage & stunning aesthetics: Excels at panoramas, infographics, and virtual try-ons, delivering realistic textures and elegant typography. Start to create your next masterpiece with Qwen-Image-2.1! šŸ–¼ļø - Blog: qwen.ai/blog?id=qwen-image-2ā€¦ - GitHub: github.com/QwenLM/Qwen-Imageā€¦ - Model Scope: modelscope.cn/models/Qwen/Qwā€¦ - Hugging Face: huggingface.co/Qwen/Qwen-Imaā€¦
76
271
34
2,907
363,213
Qwen-Image-2.1 FP8 and GGUF quants should now run properly in Unsloth Desktop! šŸ’œ Image gen and editing are both supported. GitHub: github.com/unslothai/unsloth
6
7
3
63
14,384
Load more