CASA: Cross-Attention via Self-Attention for Efficient Vision-Language Fusion
Paper
β’
2512.19535
β’
Published
β’
10
AI & ML interests
Unofficial org for community upload of Mistral's Open Source models.
Recent Activity
View all activity
nielsrΒ
submitted
a
paper
to Daily Papers
5 days ago
danielhanchenΒ
posted
an
update
5 days ago
Post
3540
You can now run GLM-4.7, the new 355B parameter SOTA model on your local device (128GB RAM).β¨
The model achieves SOTA performance on coding, agentic and chat benchmarks.
GGUF: unsloth/GLM-4.7-GGUF
Guide: https://docs.unsloth.ai/models/glm-4.7
The model achieves SOTA performance on coding, agentic and chat benchmarks.
GGUF: unsloth/GLM-4.7-GGUF
Guide: https://docs.unsloth.ai/models/glm-4.7
- 2 replies
danielhanchenΒ
posted
an
update
10 days ago
Post
2190
Google releases FunctionGemma, a new 270M parameter model that runs on just 0.5 GB RAM.β¨
Built for tool-calling, run locally on your phone at 50+ tokens/s, or fine-tune with Unsloth & deploy to your phone.
GGUF: unsloth/functiongemma-270m-it-GGUF
Docs + Notebook: https://docs.unsloth.ai/models/functiongemma
Built for tool-calling, run locally on your phone at 50+ tokens/s, or fine-tune with Unsloth & deploy to your phone.
GGUF: unsloth/functiongemma-270m-it-GGUF
Docs + Notebook: https://docs.unsloth.ai/models/functiongemma
-
- 2 replies
danielhanchenΒ
posted
an
update
13 days ago
Post
5238
NVIDIA releases Nemotron 3 Nano, a new 30B hybrid reasoning model! π₯
Has 1M context window & best in class performance for SWE-Bench, reasoning & chat. Run the MoE model locally with 24GB RAM.
GGUF: unsloth/Nemotron-3-Nano-30B-A3B-GGUF
π Step-by-step Guide: https://docs.unsloth.ai/models/nemotron-3
Has 1M context window & best in class performance for SWE-Bench, reasoning & chat. Run the MoE model locally with 24GB RAM.
GGUF: unsloth/Nemotron-3-Nano-30B-A3B-GGUF
π Step-by-step Guide: https://docs.unsloth.ai/models/nemotron-3
- 1 reply
danielhanchenΒ
posted
an
update
17 days ago
Post
1992
Mistral's new SOTA coding models Devstral 2 can now be Run locally! (25GB RAM) π±
We fixed the chat template, so performance should be much better now!
24B: unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF
123B: unsloth/Devstral-2-123B-Instruct-2512-GGUF
π§‘Step-by-step Guide: https://docs.unsloth.ai/models/devstral-2
We fixed the chat template, so performance should be much better now!
24B: unsloth/Devstral-Small-2-24B-Instruct-2512-GGUF
123B: unsloth/Devstral-2-123B-Instruct-2512-GGUF
π§‘Step-by-step Guide: https://docs.unsloth.ai/models/devstral-2
mrfakenameΒ
posted
an
update
24 days ago
Post
4167
Excited to share that I've joined the Hugging Face Fellows program! π€
Looking forward to contributing to & working more closely with the open-source ecosystem - huge thanks to everyone who's supported me on this journey! π
Looking forward to contributing to & working more closely with the open-source ecosystem - huge thanks to everyone who's supported me on this journey! π
danielhanchenΒ
posted
an
update
25 days ago
Post
3636
Mistral's new Ministral 3 models can now be Run & Fine-tuned locally! (16GB RAM)
Ministral 3 have vision support and the best-in-class performance for their sizes.
14B Instruct GGUF: unsloth/Ministral-3-14B-Instruct-2512-GGUF
14B Reasoning GGUF: unsloth/Ministral-3-14B-Reasoning-2512-GGUF
π± Step-by-step Guide: https://docs.unsloth.ai/new/ministral-3
All GGUFs, BnB, FP8 etc. variants uploads: https://huggingface.co/collections/unsloth/ministral-3
Ministral 3 have vision support and the best-in-class performance for their sizes.
14B Instruct GGUF: unsloth/Ministral-3-14B-Instruct-2512-GGUF
14B Reasoning GGUF: unsloth/Ministral-3-14B-Reasoning-2512-GGUF
π± Step-by-step Guide: https://docs.unsloth.ai/new/ministral-3
All GGUFs, BnB, FP8 etc. variants uploads: https://huggingface.co/collections/unsloth/ministral-3
-
- 3 replies
danielhanchenΒ
posted
an
update
30 days ago
Post
8445
Qwen3-Next can now be Run locally! (30GB RAM)
Instruct GGUF: unsloth/Qwen3-Next-80B-A3B-Instruct-GGUF
The models come in Thinking and Instruct versions and utilize a new architecture, allowing it to have ~10x faster inference than Qwen32B.
π Step-by-step Guide: https://docs.unsloth.ai/models/qwen3-next
Thinking GGUF: unsloth/Qwen3-Next-80B-A3B-Thinking-GGUF
Instruct GGUF: unsloth/Qwen3-Next-80B-A3B-Instruct-GGUF
The models come in Thinking and Instruct versions and utilize a new architecture, allowing it to have ~10x faster inference than Qwen32B.
π Step-by-step Guide: https://docs.unsloth.ai/models/qwen3-next
Thinking GGUF: unsloth/Qwen3-Next-80B-A3B-Thinking-GGUF
reach-vbΒ
authored
a
paper
about 1 month ago
danielhanchenΒ
posted
an
update
about 2 months ago
Post
4358
You can now run Kimi K2 Thinking locally with our Dynamic 1-bit GGUFs:
unsloth/Kimi-K2-Thinking-GGUF
We shrank the 1T model to 245GB (-62%) & retained ~85% of accuracy on Aider Polyglot. Run on >247GB RAM for fast inference.
We also collaborated with the Moonshot AI Kimi team on a system prompt fix! π₯°
Guide + fix details: https://docs.unsloth.ai/models/kimi-k2-thinking-how-to-run-locally
We shrank the 1T model to 245GB (-62%) & retained ~85% of accuracy on Aider Polyglot. Run on >247GB RAM for fast inference.
We also collaborated with the Moonshot AI Kimi team on a system prompt fix! π₯°
Guide + fix details: https://docs.unsloth.ai/models/kimi-k2-thinking-how-to-run-locally
mrfakenameΒ
posted
an
update
2 months ago
Post
6080
Trained a model for emotion-controllable TTS based on MiMo audio on LAION's dataset.
Still very early and does have an issue with hallucinating but results seem pretty good so far, given that it is very early into the training run.
Will probably kick off a new run later with some settings tweaked.
Put up a demo here: https://huggingface.co/spaces/mrfakename/EmoAct-MiMo
(Turn π on to hear audio samples)
Still very early and does have an issue with hallucinating but results seem pretty good so far, given that it is very early into the training run.
Will probably kick off a new run later with some settings tweaked.
Put up a demo here: https://huggingface.co/spaces/mrfakename/EmoAct-MiMo
(Turn π on to hear audio samples)
-
- 5 replies
JustinLin610Β
authored
a
paper
3 months ago
danielhanchenΒ
posted
an
update
4 months ago
Post
6568
Run DeepSeek-V3.1 locally on 170GB RAM with Dynamic 1-bit GGUFs!π
GGUFs: unsloth/DeepSeek-V3.1-GGUF
The 715GB model gets reduced to 170GB (-80% size) by smartly quantizing layers.
The 1-bit GGUF passes all our code tests & we fixed the chat template for llama.cpp supported backends.
Guide: https://docs.unsloth.ai/basics/deepseek-v3.1
GGUFs: unsloth/DeepSeek-V3.1-GGUF
The 715GB model gets reduced to 170GB (-80% size) by smartly quantizing layers.
The 1-bit GGUF passes all our code tests & we fixed the chat template for llama.cpp supported backends.
Guide: https://docs.unsloth.ai/basics/deepseek-v3.1
Post
4799
Thread to gossip during the
openai
GPT-5 livestream: https://www.youtube.com/watch?v=0Uu_VJeVVfo. Feel free to post your impressions below!
- 29 replies
MaziyarPanahiΒ
authored
2
papers
5 months ago
danielhanchenΒ
posted
an
update
5 months ago
Post
5633
Run OpenAI's new gpt-oss models locally with Unsloth GGUFs! π₯π¦₯
20b GGUF: unsloth/gpt-oss-20b-GGUF
120b GGUF: unsloth/gpt-oss-120b-GGUF
Model will run on 14GB RAM for 20b and 66GB for 120b.
20b GGUF: unsloth/gpt-oss-20b-GGUF
120b GGUF: unsloth/gpt-oss-120b-GGUF
Model will run on 14GB RAM for 20b and 66GB for 120b.
- 2 replies
JustinLin610Β
authored
a
paper
5 months ago
danielhanchenΒ
posted
an
update
5 months ago
Post
3737
It's Qwen3 week! π We uploaded Dynamic 2-bit GGUFs for:
Qwen3-Coder: unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF
Qwen3-2507: unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF
So you can run them both locally!
Guides are in model cards.
Qwen3-Coder: unsloth/Qwen3-Coder-480B-A35B-Instruct-GGUF
Qwen3-2507: unsloth/Qwen3-235B-A22B-Instruct-2507-GGUF
So you can run them both locally!
Guides are in model cards.
- 1 reply
MaziyarPanahiΒ
posted
an
update
5 months ago
Post
12666
𧬠Breaking news in Clinical AI: Introducing the OpenMed NER Model Discovery App on Hugging Face π¬
OpenMed is back! π₯ Finding the right biomedical NER model just became as precise as a PCR assay!
I'm thrilled to unveil my comprehensive OpenMed Named Entity Recognition Model Discovery App that puts 384 specialized biomedical AI models at your fingertips.
π― Why This Matters in Healthcare AI:
Traditional clinical text mining required hours of manual model evaluation. My Discovery App instantly connects researchers, clinicians, and data scientists with the exact NER models they need for their biomedical entity extraction tasks.
π¬ What You Can Discover:
β Pharmacological Models - Extract "chemical compounds", "drug interactions", and "pharmaceutical" entities from clinical notes
β Genomics & Proteomics - Identify "DNA sequences", "RNA transcripts", "gene variants", "protein complexes", and "cell lines"
β Pathology & Disease Detection - Recognize "pathological formations", "cancer types", and "disease entities" in medical literature
β Anatomical Recognition - Map "anatomical systems", "tissue types", "organ structures", and "cellular components"
β Clinical Entity Extraction - Detect "organism species", "amino acids", 'protein families", and "multi-tissue structures"
π‘ Advanced Features:
π Intelligent Entity Search - Find models by specific biomedical entities (e.g., "Show me models detecting CHEM + DNA + Protein")
π₯ Domain-Specific Filtering - Browse by Oncology, Pharmacology, Genomics, Pathology, Hematology, and more
π Model Architecture Insights - Compare BERT, RoBERTa, and DeBERTa implementations
β‘ Real-Time Search - Auto-filtering as you type, no search buttons needed
π¨ Clinical-Grade UI - Beautiful, intuitive interface designed for medical professionals
Ready to revolutionize your biomedical NLP pipeline?
π Try it now: OpenMed/openmed-ner-models
𧬠Built with: Gradio, Transformers, Advanced Entity Mapping
OpenMed is back! π₯ Finding the right biomedical NER model just became as precise as a PCR assay!
I'm thrilled to unveil my comprehensive OpenMed Named Entity Recognition Model Discovery App that puts 384 specialized biomedical AI models at your fingertips.
π― Why This Matters in Healthcare AI:
Traditional clinical text mining required hours of manual model evaluation. My Discovery App instantly connects researchers, clinicians, and data scientists with the exact NER models they need for their biomedical entity extraction tasks.
π¬ What You Can Discover:
β Pharmacological Models - Extract "chemical compounds", "drug interactions", and "pharmaceutical" entities from clinical notes
β Genomics & Proteomics - Identify "DNA sequences", "RNA transcripts", "gene variants", "protein complexes", and "cell lines"
β Pathology & Disease Detection - Recognize "pathological formations", "cancer types", and "disease entities" in medical literature
β Anatomical Recognition - Map "anatomical systems", "tissue types", "organ structures", and "cellular components"
β Clinical Entity Extraction - Detect "organism species", "amino acids", 'protein families", and "multi-tissue structures"
π‘ Advanced Features:
π Intelligent Entity Search - Find models by specific biomedical entities (e.g., "Show me models detecting CHEM + DNA + Protein")
π₯ Domain-Specific Filtering - Browse by Oncology, Pharmacology, Genomics, Pathology, Hematology, and more
π Model Architecture Insights - Compare BERT, RoBERTa, and DeBERTa implementations
β‘ Real-Time Search - Auto-filtering as you type, no search buttons needed
π¨ Clinical-Grade UI - Beautiful, intuitive interface designed for medical professionals
Ready to revolutionize your biomedical NLP pipeline?
π Try it now: OpenMed/openmed-ner-models
𧬠Built with: Gradio, Transformers, Advanced Entity Mapping
-
- 5 replies