Downloads · 30 days
14
18% of all-time downloads
pplx-embed-v1-0.6b-onnx-int8-server
tss-deposium/pplx-embed-v1-0.6b-onnx-int8-server
pplx-embed-v1-0.6b-onnx-int8-server is a feature extraction model from tss-deposium. Use it when you need embeddings to search or compare text. It is set up for onnx. The card lists the license as mit.
This repository contains the exact server-side artifact used by Deposium deposium-turbov3 for pplx-embed-v1 embedding inference.
All-time downloads
78
Public
Repo size
718 MB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.onnx_data706 MB · 98%
At a glance
- Task
- Feature Extraction
- Library
- onnx
- License
- mit
- Model type
- bidirectional_pplx_qwen3
- Access
- Public
- Created
- Jun 22, 2026
- Updated
- Jun 22, 2026
- SHA
- d37df300
Model card
From the Hugging Face model README
PPLX Embed v1 0.6B ONNX INT8 server artifact
This repository contains the exact server-side artifact used by Deposium
deposium-turbov3 for pplx-embed-v1 embedding inference.
It is not the browser/Transformers.js single-file export. It uses ONNX external data:
model_quantized.onnxmodel_quantized.onnx_dataconfig.jsontokenizer.jsontokenizer_config.jsonspecial_tokens_map.json
Runtime contract observed in turbov3:
- registered model name:
pplx-embed-v1 - architecture:
PPLXQwen3Model - model type:
bidirectional_pplx_qwen3 - output used by server:
pooler_output - native output dimension: 1024
- production consumer dimension: 256 via Matryoshka truncation
- max context: 32768 tokens
Checksums from the source container:
4f8df41878c7a8cb483516eac180aa61843196e3dd7ec36cc0e83c79abc2047f model_quantized.onnx
4d7f80998954f14d9555e6655204d928c35f5477499f95acef3002e7eec62b91 model_quantized.onnx_data
f7865547ecc1c077c5b1f83913fb9816a6586c676ccf3e0f7f00105a1b0c1c6c config.json
c6fb5c5bbba5fa5f8332edfb6d8aa67bd7fb3d75365b1765f108201698eaebf5 tokenizer.json
- Task
- Feature Extraction
- Library
- onnx
- Type
- bidirectional_pplx_qwen3
- License
- mit
- Created
- Jun 22, 2026
- Updated
- Jun 22, 2026