Downloads · 30 days
15
4% of all-time downloads
bge-reasoner-embed-qwen3-8b-0923-AWQ-4bit
sionic-ai/bge-reasoner-embed-qwen3-8b-0923-AWQ-4bit
bge-reasoner-embed-qwen3-8b-0923-AWQ-4bit is a text ranking model from sionic-ai. Use it for the text ranking 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.
All-time downloads
352
Public
Parameters
8.2B
8.6 GB on disk
Likes
6
Public
Repo makeup
Click a slice to open those files.
.safetensors8.6 GB · 100%
Parameter types
How the weights are stored.
I326.9B · 85%
Base models
Model card
From the Hugging Face model README
Evaluations
[EMB] AWQ: encoding corpus ...
corpus time=121.40s | 42.69 docs/s
[EMB] AWQ: encoding queries ...
query time=4.12s | 72.81 q/s
[RET] Ranking top-10 ...
==================== Tiny-BEIR (SciFact/test) ====================
Encoding speed (corpus | queries):
Base : 26.62 docs/s | 106.20 q/s
AWQ : 42.69 docs/s | 72.81 q/s
Metrics (nDCG@10 / Recall@1 / Recall@10 / MRR@10):
Base : 73.25% | 55.63% | 88.66% | 68.53%
AWQ : 73.43% | 58.13% | 86.99% | 69.36%
Evaluated queries: 300
===============================================================
- Task
- Text Ranking
- Type
- qwen3
- License
- apache-2.0
- Created
- Sep 27, 2025
- Updated
- Sep 28, 2025