List view
For now, this is a milestone to track running progress of the Runtime through FEs and MLIR.
No due date•3/23 issues closed- Due by December 31, 2025•41/150 issues closed
- No due date•3/7 issues closed
- Due by November 30, 2025•2/9 issues closed
- Overdue by 1 month(s)•Due by September 30, 2025
- Overdue by 1 month(s)•Due by September 30, 2025•2/4 issues closed
MLIR related milestone: https://github.com/tenstorrent/tt-mlir/milestone/50
Overdue by 26 day(s)•Due by October 31, 2025MLIR Related milestone: https://github.com/tenstorrent/tt-mlir/milestone/46
Overdue by 26 day(s)•Due by October 31, 202530 of our top internal models at 10% of top perf. MLIR Related milestone: https://github.com/tenstorrent/tt-mlir/milestone/45
Overdue by 26 day(s)•Due by October 31, 2025•8/18 issues closedThrough tt-xla we need to support EmitPy. MLIR equivalent: https://github.com/tenstorrent/tt-mlir/milestone/47
Overdue by 1 month(s)•Due by September 30, 2025We want the ability to start supporting multi-host scaleout. For example being able to combine multiple 1,8 mesh sized machines e.g. 4 LoudBoxes [4,8] mesh size.
Due by December 31, 2025•2/3 issues closedFrom our internal priority list of Models we need to have 50 of these running end to end.
Overdue by 26 day(s)•Due by October 31, 2025•2/52 issues closedTracking bringup of priority customer models through tt-xla frontend.
No due date•0/3 issues closed