Downloads · 30 days
14
11% of all-time downloads
Marco-Nano-Instruct-REAP-7B-A0.6B
sasa2000/Marco-Nano-Instruct-REAP-7B-A0.6B
Marco-Nano-Instruct-REAP-7B-A0.6B is a text generation model from sasa2000. Use it when you need the model to write or continue text. It is set up for transformers. The card lists the license as apache-2.0.
This is a pruned derivative of ATH-MaaS/Marco-Nano-Instruct, produced with CerebrasResearch/reap layerwise REAP pruning in a Codex-assisted experiment (GPT5.5 xhigh ).
All-time downloads
124
Public
Parameters
6.9B
13.8 GB on disk
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors13.8 GB · 100%
At a glance
- Task
- Text Generation
- Library
- transformers
- License
- apache-2.0
- Model type
- qwen3_moe
- Access
- Public
- Created
- Jun 27, 2026
- Updated
- Jun 30, 2026
- SHA
- 37af09ac
Try a prompt
Base models
Model card
From the Hugging Face model README
Marco-Nano-Instruct REAP Pruned 0.15
This is a pruned derivative of ATH-MaaS/Marco-Nano-Instruct, produced with
CerebrasResearch/reap layerwise REAP pruning in a Codex-assisted experiment
(GPT5.5 xhigh ).
Details
- Pruning method: REAP layerwise pruning
- Requested compression ratio:
0.15 - Experts retained:
198 - Experts per token:
8 - Calibration dataset:
theblackcat102/evol-codealpaca-v1 model_max_length:2048batches_per_category:128batch_size:1batch_group_size:8- Format: safetensors
Notes
This checkpoint was created for a low-resource pruning experiment on a 6GB VRAM environment. Benchmark evaluation was not run. A short prompt smoke check showed this ratio to be the most stable of the tested Marco-Nano pruning ratios, especially in English; Japanese and Chinese outputs may still show repetition or wording artifacts.
- Task
- Text Generation
- Library
- transformers
- Type
- qwen3_moe
- License
- apache-2.0
- Created
- Jun 27, 2026
- Updated
- Jun 30, 2026