Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
NullSense
/
Nanbeige4.2-3B-FP8-Dynamic
like
0
Text Generation
Safetensors
English
Chinese
nanbeige
fp8
nvfp4
quantized
compressed-tensors
vllm
looped-transformer
conversational
custom_code
Eval Results (legacy)
arxiv:
2605.16343
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
Nanbeige4.2-3B-FP8-Dynamic
/
assets
Ctrl+K
Ctrl+K
1 contributor
History:
5 commits
NullSense
Refresh EAGLE-3 references for the thinking-aware v2 head (2026-07-24 retrain); concurrency section re-measured (501/878/1017 tok/s, ITL 6.2-7.6ms)
4b8767c
verified
4 days ago
chart_concurrency.png
Safe
49.3 kB
xet
Refresh EAGLE-3 references for the thinking-aware v2 head (2026-07-24 retrain); concurrency section re-measured (501/878/1017 tok/s, ITL 6.2-7.6ms)
4 days ago
chart_quality_benches.png
Safe
135 kB
xet
Add multi-bench quality comparison chart
5 days ago
chart_quality_vs_size.png
Safe
87.9 kB
Add family charts (quality-vs-size, speed per workload)
5 days ago
chart_speed_workloads.png
Safe
69.3 kB
Add family charts (quality-vs-size, speed per workload)
5 days ago
