Tag
A community fine-tuned and merged version of Qwen 3.5 9B that claims to exceed critical benchmarks of larger models, available in GGUF format with multi-token prediction support.
DavidAU releases Qwen3.6-27B-Fable-Fusion-711, a multi-stage fine-tune of Qwen 3.6 27B that claims to exceed 700 ARC-C benchmark, surpassing base models and matching closed-source models, available in GGUF format for consumer hardware.
A community fine-tune of Krea 2 Raw that enables instruction-based, identity-preserving image editing. It edits images while preserving details like faces, using a custom ComfyUI node pack.
Grug-35B-A3B is an open-source experimental model based on Qwen-3.6-35B-A3B that reduces reasoning token usage by approximately 69.8% while maintaining answer quality within 2% of the base model, resulting in faster generation and reduced context usage.
A focused fine-tune of Gemma 4 12B for coding, distilled from chain-of-thought data (Composer 2.5 and Fable 5) and quantized to GGUF for local, offline use with minimal VRAM requirements.
Super Gemma 4 26B Uncensored GGUF v2 is a community fine-tuned model offering uncensored responses with zero refusals, improved speed, and fixed tool-calling, optimized for local inference on llama.cpp and vLLM.
H Company releases Holo3.1, a family of Vision-Language Models (0.8B to 35B) for computer use agents, supporting web, desktop, and mobile automation with native function calling and optimized quantized checkpoints for local deployment.
The MiniMax M2.7 model has been fine-tuned into an uncensored variant, 'ultra uncensored heretic', with very low refusal rates (4/100). Available in Safetensors and GGUF formats on HuggingFace.
Dramabox is an expressive text-to-speech model by Resemble AI that uses prompt-driven control for speaker identity, emotion, and delivery, with optional voice cloning via a 10-second reference. Built on the LTX-2.3 audio diffusion transformer, it is open-sourced on Hugging Face.