A new nightly version has been released!
See the quickstart guide for installation instructions: Quickstart | Modular
MAX changelog updates:
- Fixed a rare illegal-instruction crash in the SM100 (Blackwell)
flash-attention prefill kernels under chunked prefill with tensor
parallelism. When the attention grid shared SMs with the tensor-parallel
all-reduce collective under device graph capture, a consumer warp could read
a stale tensor-memory base address and issue a tensor-core MMA against an
invalid operand. The kernels now read the tensor-memory base once after it
is published and carry it in a register, so there is no in-loop re-read to
race.
Mojo changelog updates:
- [stdlib] Deprecate
UnsafePointerimplicit conversion toUnsafeAnyOrigin - [stdlib] Add
warp.match_allGPU primitive
Raw MAX diff: https://github.com/modular/modular/compare/c9aebe6b32e63539eff6366392501e2724af6070...c3a013405ada7137a5bb81c2540b978cf0151e26)>
Current Mojo changelog: https://github.com/modular/modular/blob/main/mojo/docs/nightly-changelog.md