Hugging Face
Search models, datasets, users...
Ctrl K
Models
Datasets
Spaces
Buckets
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
Spaces:
devon7y
/
rat-qwen
Like
0
Running
on
Zero
App
Files
Files
Community
Fetching metadata from the HF Docker repository...
main
rat-qwen
632 kB
Ctrl+K
Ctrl+K
1 contributor
History:
5 commits
devon7y
Warm FP8/linear-attention kernels once at startup
d34150c
verified
20 days ago
validation
FP8 vs bf16 dose-response check: FP8 reproduces the effect at every dose
20 days ago
.gitattributes
Safe
1.52 kB
initial commit
20 days ago
README.md
Safe
1.13 kB
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
app.py
Safe
10.8 kB
Warm FP8/linear-attention kernels once at startup
20 days ago
index.html
Safe
18 kB
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
make_rat_vector.py
Safe
964 Bytes
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
ocd_core.py
Safe
23.3 kB
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
rat_vector.pt
22.2 kB
xet
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
requirements.txt
Safe
100 Bytes
Rat-Qwen on ZeroGPU: Qwen3.5-27B-FP8, L51 f26585
20 days ago
validate_fp8.py
Safe
3.79 kB
Persist Triton autotune; size GPU durations from measured speed; one padded batch per validation dose
20 days ago
