Downloads · 30 days
14
33% of all-time downloads
Qwen3.8-27B-int4-asym-archive
SHACortes/Qwen3.8-27B-int4-asym-archive
Qwen3.8-27B-int4-asym-archive is a machine learning model from SHACortes. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as apache-2.0.
This is a broken export, kept for reference only.
All-time downloads
43
Public
Repo size
21.9 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin21.9 GB · 100%
At a glance
- License
- apache-2.0
- Model type
- qwen3_5
- Access
- Public
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026
- SHA
- d0dae651
Model card
From the Hugging Face model README
Qwen3.8-27B int4 asym (ARCHIVED — do not use)
This is a broken export, kept for reference only.
Why it's archived
Asymmetric (int4_asym) weight compression on this model causes a floating-point
exception (FPE) crash when running tool-call prompts on Arc B70
(OpenVINO GenAI). In short: ASYM quantization is unreliable at this
compression ratio for this model/hardware combo.
Tracked upstream: https://github.com/openvinotoolkit/openvino.genai/issues/3870
Replacement
Use the symmetric (int4_sym) export instead — same calibration setup,
no crash: Qwen3.6-27B-r06-sym-no-mtp / Qwen3.8-27B-int4-sym in the
original export directory.
Contents
OpenVINO IR export (language model, text/vision embeddings, tokenizer, detokenizer) for a Qwen3-VL 27B model, weight-compressed with mixed int4_asym/int8_asym precision (57% int4_asym / 43% int8_asym by parameter count).
- Type
- qwen3_5
- License
- apache-2.0
- Created
- Aug 18, 2026
- Updated
- Aug 18, 2026