---
license: apache-2.0
language:
- en
base_model:
- Qwen/Qwen2.5-Coder-7B
- open-r1/OlympicCoder-7B
pipeline_tag: text-generation
tags:
- merge
- programming
- code generation
- code
- qwen2
- codeqwen
- chat
- qwen
- qwen-coder
---

<h2>Qwen2.5-2X11B-CODER-Dueling-Wolverines-25B-gguf</h2>

<img src="duel-wolverine-coder.jpg" style="float:right; width:300px; height:500px; padding:10px;">

"Ripping your programming worries to shreds... fast."

Tipping the scales at 42 layers and 507 tensors... the monster lives.

Two monsters in fact - in one.

This is MOE model, using V1 and V2 of Wolverine-Coder 11B which is a merge of two models noted below.

The MOE config gives you full access to both 11B models at full power.

This MOE model generates stronger, more compact code with an enhanced understanding of your instructions 
and follows what you tell them to the letter.

Each 11B version is an overpowered - yet wickedly fast - CODING ENGINE are based on two of the best coder AIs:

"Qwen2.5-Coder-7B-Instruct" 

[ https://huggingface.co/Qwen/Qwen2.5-Coder-7B-Instruct ]

and

"OlympicCoder-7B"

[ https://huggingface.co/open-r1/OlympicCoder-7B ]

These two models are stuffed into one compact powerhouse 11BX2 merge that is stronger in performance and understanding than both donor models.

There are TWO versions of this MOE model.

Quants Q3_K_M and Q6_K are available at the moment, of each version.

These are generated from bfloat16 source files.

Final models will be generated from float32 source files, to improve performance of the MOE model further.

NOTES: 
- Each config/version will be very different from each other.
- You can select 1 or 2 experts, default is 2 experts.
- Due to unique setup of this moe, suggest 1-4 generations.
- Tool Calling is supported in both versions.
- Source(s) / full quanting to follow // full repos to follow.
- Final model size (including layers/tensors) / config subject to change.

---

Config / Settings

---

Model is set at 32k/32768 context for these GGUFS, full quants/full repos will be 128k/131072.

Requirements [Qwen 2.5 7B Coder default settings]:
- Temp .5 to .7 (or lower)
- topk: 20, topp: .8, minp: .05
- rep pen: 1.1 (can be lower)
- Jinja Template (embedded) or CHATML template.
- A System Prompt is not required. (ran tests with blank system prompt)

Refer to either "Qwen2.5-Coder-7B-Instruct" and/or "OlympicCoder-7B" repos (above) for additional settings, benchmarks and usage.

---