We're sharing/showcasing best of @huggingface models. Follow to stay in loop. Promoting Open-Source models.
- Tweets 29,723
- Following 32
- Followers 68,765
- Likes 567
Pinned Tweet
Make a good career out of your GitHub repo.
It’s happening…
Your GitHub is just not just a repo, but a goldmine for you.
Keep your open-source projects ready.
The winter is coming…
opencontest.dev
1
24
9,918
Ever wanted a 27B model that runs on almost nothing? Meet Ternary Bonsai 2: a GGUF build squeezed into 2-bit ternary weights. Tiny footprint, huge brain energy. Perfect for local setups.
1
1
5
418
Meet Swift-1.5-Qwen3.8-27B-GSQ-RCO-GGUF: a 27B parameter model built on Qwen3.8, optimized for reasoning and token efficiency. With 22k+ downloads, it's making waves for its smart thinking and GGUF format.
1
2
606
Ever wished AI could write Nabati poetry that feels truly Saudi? Mawrooth-ALLaM-7B-LoRA is here! This 7B model fine-tuned with LoRA on Arabic poetry brings cultural heritage to life. A must-try for Arabic NLP fans.
3
6
958
Built on google/gemma-4-31B with 31B parameters. It uses the transformers library and safetensors format. The model card references an arXiv paper (2607.02770) and includes eval results, so you can check benchmarks before deploying.
1
123
Meet gemma-4-31B-it-custom: a 31B image-text-to-text model fine-tuned from Google's Gemma 4. It handles conversations with images and text, all under Apache 2.0. A solid pick for multimodal builders.
1
9
1,391