anik-jha/Qwen3.6-35B-A3B-coding-reap50-GGUF overview
Qwen3.6 35B A3B coding specialist, 50% experts pruned GGUF Qwen3.6 35B A3B with half of its experts removed 128 of 256 per layer, REAP scoring on a coding heav…
Runs locally from ~10.59 GB disk (12 GB VRAM class GPUs with llama.cpp / guIDE).
Repository Files & Downloads
Model Details
Model README
---
license: apache-2.0
base_model: Qwen/Qwen3.6-35B-A3B
tags:
- gguf
- mixture-of-experts
- pruning
- code
library_name: llama.cpp
---
Qwen3.6-35B-A3B coding specialist, 50% experts pruned (GGUF)
Qwen3.6-35B-A3B with half of its experts removed (128 of 256 per layer, REAP
scoring on a coding-heavy calibration mix), quantized with an importance
matrix. This is the main artifact of the paper *Half the Experts, All the
Code* (arXiv link coming): a 19B-parameter coding specialist that fits where
the full model does not.
Why you might want it: at the same memory budget, these beat 2-bit
quantization of the full model on HumanEval+.
| file | size | HumanEval+ | MBPP+ |
|---|---|---|---|
| Q4_K_M (imatrix) | 11.4 GB | 0.902 | 0.720 |
| Q5_K_M (imatrix) | 13.3 GB | 0.915 | 0.749 |
| full model, IQ2_M (for comparison, not included) | 13.0 GB | 0.896 | 0.730 |
| full model, Q8 (for comparison, not included) | 38 GB | 0.890 | 0.772 |
All numbers are greedy pass@1 via EvalPlus against llama.cpp, reasoning off.
The trade is real: general (non-coding) perplexity roughly doubles. This is a
coding specialist, not a general assistant - don't deploy it as one.
Made with moep; the exact expert
selection JSON is in that repo under artifacts/selections/, so you can
reproduce the surgery from the base checkpoint. Runs anywhere llama.cpp
runs.
Derivative of Qwen3.6-35B-A3B, Apache-2.0, upstream notice retained.
Run anik-jha/Qwen3.6-35B-A3B-coding-reap50-GGUF with guIDE
Download guIDE — the AI-native code editor with local LLM inference and 69 built-in tools.
Source: Hugging Face · Compare models