Huihui-Qwen3.6-35B-A3B-abliterated-MTP
Qwen/Qwen3.6-35B-A3B
Highly popular MoE abliterated variant. Excellent quality-to-size ratio for the active parameter count.
- Approx. VRAM
- ~21 GB (Q4)
- License
- Apache-2.0
- Downloads
- 5.2M / 30d
updated Nov 15, 2026
Search models, news, guides and FAQs.
ConwayResearch/Underdog-Saluki-27B-1.0
A 27B abliterated build with MTP/speculative decoding support, tuned for tool use and agent workflows. The base model is non-commercial (CC-BY-NC-4.0).
This repository ships GGUF files. Ollama can pull them from Hugging Face; adjust the quant tag if needed:
ollama run hf.co/Kujira/Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF:Q4_K_MUses the Hugging Face CLI. Install it with pip install -U huggingface_hub if needed.
hf download Kujira/Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUFIn LM Studio, search for the repository name and pick a GGUF quant that fits your VRAM. With llama.cpp, use llama-cli -hf Kujira/Underdog-Saluki-27B-1.0-MTP-Abliterated-GGUF. See the setup guide for full flags and sampling defaults.
Listed quantizations: GGUF.
A 27B abliterated build with MTP/speculative decoding support, tuned for tool use and agent workflows. The base model is non-commercial (CC-BY-NC-4.0).
Method: abliteration. License: CC-BY-NC-4.0. Last catalog update: Oct 9, 2026. This entry links to a third-party repository; wahloon.net is not affiliated with the publisher. Verify the current model card, license and file list before downloading. You are responsible for how you use the model.
Qwen/Qwen3.6-35B-A3B
Highly popular MoE abliterated variant. Excellent quality-to-size ratio for the active parameter count.
updated Nov 15, 2026
Qwen/Qwen3.8-27B
One of the most downloaded abliterated Qwen 3.8 27B builds. Strong general performance with significantly reduced refusals.
Qwen/Qwen3.8-Flash-Next
Very high download count Flash-Next abliterated build. Strong for longer context and throughput-oriented use.
updated Nov 10, 2026