A new nightly version has been released! ![]()
See the quickstart guide for installation instructions: Quickstart | Modular
MAX changelog updates:
- Fixed an expert-parallelism dispatch assertion (
Cannot dispatch EP kernel with N input tokens when the maximum tokens per rank is N-1)
that fired whenever--max-batch-input-tokenswas not evenly
divisible by the tensor-parallel degree. The EP per-rank cap now uses
ceiling division to match the ragged binning ofreducescatterin
TP-attention + EP-MoE mode, so the largest shard fits in the
dispatch buffer. Affects DeepSeek-V3, Kimi-K2.5, MiniMax-M2, Qwen3,
and Step3.5 deployments configured with non-divisible batch sizes.
Mojo changelog updates:
- [stdlib] Drop
ImplicitlyDestructiblefromImplicitlyCopyabletrait - [mojo] Print decls in diags with unreadable locs
Raw MAX diff: https://github.com/modular/modular/compare/29a2f54c7b44e5548a5827a76528fb2bf73c9888...2fc2023f9f524ce031df72e92b195669f41033c1)>
Current Mojo changelog: https://github.com/modular/modular/blob/main/mojo/docs/nightly-changelog.md