Install this package:
emerge -a dev-python/tokenspeed-mla-bin
<pkgmetadata> <maintainer type="person"> <email>iohann.s.titov@gmail.com</email> <name>Ivan S. Titov</name> </maintainer> <longdescription>TokenSpeed MLA provides precompiled and just-in-time CUDA kernels for multi-head latent attention on NVIDIA Blackwell GPUs.</longdescription> <upstream> <remote-id type="pypi">tokenspeed-mla</remote-id> </upstream> </pkgmetadata>
| Type | File | Size | Versions |
|---|---|---|---|
| DIST | tokenspeed_mla-0.1.8-py3-none-manylinux_2_28_x86_64.whl | 755827 bytes | 0.1.8 |
| DIST | tokenspeed_mla-0.2.5-py3-none-manylinux_2_28_x86_64.whl | 755951 bytes | 0.2.5 |
| Type | File | Size |
|---|