125 Pulls 1 Tag Updated 14 hours ago
Decensored SmolLM3-3B processed with Heretic abliteration. Achieves 63/100 refusals with an exceptionally low KL divergence of 0.0001 — near-perfect model quality preservation on a consumer RTX 4060. GGUF format, ready to run.
60 Pulls 1 Tag Updated 1 week ago
Fully decensored Qwen2.5-3B-Instruct processed with Heretic abliteration. Achieves 3/100 refusals with 0.11 KL divergence — 97% censorship removal on a consumer RTX 4060. GGUF format, ready to run.
238 Pulls 1 Tag Updated 1 week ago
1,981 Pulls 2 Tags Updated 2 weeks ago
213 Pulls 1 Tag Updated 4 weeks ago
1,093 Pulls 1 Tag Updated 4 weeks ago
239 Pulls 1 Tag Updated 1 month ago
3,831 Pulls 1 Tag Updated 1 month ago
10.6K Pulls 1 Tag Updated 1 month ago
Privacidad, Rapidez, Muy bueno en todo incluido en agentes y sin censura
395 Pulls 1 Tag Updated 1 month ago
[El mejor 03-03-2026] privacidad absoluta, velocidad extrema, cero censura, nivel doctorado.
1,491 Pulls 1 Tag Updated 1 month ago
270 Pulls 1 Tag Updated 1 month ago
Qwen 3.5 is a family of open-source multimodal models that delivers exceptional utility and performance.
106.6K Pulls 50 Tags Updated 5 hours ago
Qwen3-Coder-Next is a coding-focused language model from Alibaba's Qwen team, optimized for agentic coding workflows and local development.
4,633 Pulls 4 Tags Updated 1 month ago
1,100 Pulls 1 Tag Updated 2 months ago
394 Pulls 1 Tag Updated 2 months ago
Athenea-4B-Coding is a fine-tuned version of huihui-ai/Huihui-Qwen3-4B-Thinking-2507-abliterated, specialized in code reasoning, debugging, and problem solving.
344 Pulls 1 Tag Updated 2 months ago
212 Pulls 1 Tag Updated 2 months ago
Advancing the Coding Capability
611 Pulls 1 Tag Updated 2 months ago
LFM2.5 is a new family of hybrid models designed for on-device deployment.
2,635 Pulls 10 Tags Updated 2 months ago