tlc-pack / libflash_attn Goto Github PK
View Code? Open in Web Editor NEWStandalone Flash Attention v2 kernel without libtorch dependency
License: BSD 3-Clause "New" or "Revised" License
Standalone Flash Attention v2 kernel without libtorch dependency
License: BSD 3-Clause "New" or "Revised" License
$SRC_DIR/third_party/flash-attention/csrc/flash_attn/src/flash_fwd_launch_template.h(68): error: expression must have a constant value
Note #2767-D: the value of *this cannot be used as a constant
detected during instantiation of "void run_mha_fwd_hdim64<T>(Flash_fwd_params &, cudaStream_t) [with T=cutlass::half_t]"
$SRC_DIR/third_party/flash-attention/csrc/flash_attn/src/flash_fwd_hdim64_fp16_sm80.cu(25): here
Do you know how we can integrate flash attn v2 into mlc-llm?
I'm currently working on optimizing inference performance based on mlc-llm and am considering replacing the primfuncs for attention with flash attn v2. Any possible ideas for implementation?
Hi friend:
did you see flash-attention is currently adding flash-decoding. do you want to add that support in your repo?
could you please write a demo main to test the interface?
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.