Downloads · 30 days
0
TinoBruno/moecher-qwen-3.8-27b-vision-13g
moecher-qwen-3.8-27b-vision-13g is a text generation model from TinoBruno. Use it when you need the model to write or continue text. The card lists the license as apache-2.0.
This repository provides the optimized Qwen 3.8 27B Vision (13G Mode) weights formatted specifically for the MinnieTheMoEcher high-performance inference engine.
Downloads · 30 days
0
Access
Public
Updated Sep 18, 2026
Repo size
14.8 GB
Likes
0
Public
Click a slice to open those files.
.bin14.8 GB · 100%
From the Hugging Face model README
This repository provides the optimized Qwen 3.8 27B Vision (13G Mode) weights formatted specifically for the MinnieTheMoEcher high-performance inference engine.
attention_dense_layers.bin: Quantized dense weights, vision transformer, and MTP layers (~13.2 GiB).draft_lm_head_int8_bf16.bin & draft_lm_head_int8.bin: Quantized MTP draft head.draft_vocab_ids.bin, draft_vocab_ids.json, draft_vocab_ids_mapping.json, draft_vocab_ids_reverse.bin: Compact MTP draft vocabulary index maps.moecher_manifest.json: Architecture, layer configuration, and memory placement metadata.tokenizer.json: Tokenizer configuration.Run directly using the Moecher inference engine:
moecher.exe --manifest models\qwen3_8_27b_vision_13g\moecher_manifest.json --max-vram 0 --quiet
Or download automatically using the Moecher Windows Installer.