Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
adasdadsd
/
smallq-flash-attention-ascend
like
0
Chinese
English
ascend
ascendc
npu
attention
flash-attention
speculative-decoding
cann
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
Copy to bucket
new
main
smallq-flash-attention-ascend
/
tests
11.8 kB
Ctrl+K
Ctrl+K
1 contributor
History:
14 commits
adasdadsd
Replace with cleaned vector-only implementation
86d4a3b
verified
about 2 months ago
CMakeLists.txt
Safe
708 Bytes
v2.0: Full float32 precision pipeline, headDim=64 support, 448/448 production tests passed
about 2 months ago
run_model_tests.py
5.41 kB
Replace with cleaned vector-only implementation
about 2 months ago
test_aclnn.cpp
5.65 kB
Replace with cleaned vector-only implementation
about 2 months ago
