Downloads · 30 days
246
36% of all-time downloads
MiMo-V2.6-Distill-Qwen-9B-Q8_0-MTP-GGUF
IHaveNoClueAndIMustPost/MiMo-V2.6-Distill-Qwen-9B-Q8_0-MTP-GGUF
MiMo-V2.6-Distill-Qwen-9B-Q8_0-MTP-GGUF is a machine learning model from IHaveNoClueAndIMustPost. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
Patched with the MTP draft tensors and the recommended amended jinja template from the ggml-org model page.<br <brUnfortuntely neither the fixed or original Qwen 3.5 9B template seems to prevent it from loops and sub-…
All-time downloads
690
Public
Repo size
9.8 GB
Likes
3
Public
Repo makeup
Click a slice to open those files.
.gguf9.8 GB · 100%
At a glance
- Access
- Public
- Created
- Sep 22, 2026
- Updated
- Sep 24, 2026
- SHA
- 1903a8f5
Base models
Model card
From the Hugging Face model README
Patched with the MTP draft tensors and the recommended amended jinja template from the ggml-org model page.<br> <br>Unfortuntely neither the fixed or original Qwen 3.5 9B template seems to prevent it from loops and sub-par perfomance in nearly all my go-to tests. <br>For agentic work with tools I highly recommend running it tightly sandboxed on non-vital data first, making sure it works well in your setup, before doing any serious work. Or possibly just wait for official fixes or the RL version instead. <br><br> EDIT: According to online feedback the froggeric template might work better/less worse than the other alternatives. <br>Overriding templates can be loaded from file in llama.cpp by using the <code>"--chat-template-file"</code> switch.<br><br> <b>CREDITS</b><br> ggml-org - Model<br> unsloth - MTP donor model<br> Coder543 - Jinja fix<br> buzz - Grafting script<br> Qwen-chan - Doing the actual work
- Created
- Sep 22, 2026
- Updated
- Sep 24, 2026