Mase/NewComputeBench/PLENA/../ repo Separation #292
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Repo Separation - Task List
Updated Goal (2 Feb 2026)
We will create five independent GitHub repositories, each responsible for a minimal, well-defined functional unit, with no redundant or overlapping functionality between them.
1. MASE: Provides core compiler functionality, including quantization, PLENA-based passes, and other necessary passes.
2. MASE Acceleration Gen (could be a different name): Imports
mase.passesand independently maintains its own pass set, effectivelymase.passes + own.passes.3. PLENA: Defines the single-core, full-stack NPU design flow. Imports
mase.passesis necessary.4. PLENA-Dist (could be a different name): Implements system-level multi-chip interconnection simulation based on PLENA.
5. NewComputeBench: Imports
mase.passesand updates the previous fine-tuning workflow to a pre-training workflow.MASE Acceleration Gen (new repo build @firemountain154B ):
import maseif dependentMASE (integrate quantization @JiaYi):
MASE (remove overlap @Binglei):
PLENA (@GeorgeWu1204 @firemountain154B ):
import masePLENA-Dist (@przemekff ):
NewComputeBench (@JeffreyWong20 @ChengZhang-98 @przemekff @firemountain154B @GeorgeWu1204 @Binglei ):
Follow the previous task goal below
Previous Goal (1 Nov 2025)
We would like to add all NewComputeBench related code change in a single PR.
Core Module Transformation Passes
For each module-level transformation pass on New Compute Paradigm, complete the following:
SNN pass (SpikeZIP-TF @JeffreyWong20 ):
docs/tutorials/newcompute)docs/source/modules/documentation)SNN pass (STA @GeorgeWu1204 ):
docs/tutorials/newcompute)docs/source/modules/documentation)ONN pass (STA @ChengZhang-98):
docs/tutorials/newcompute)docs/source/modules/documentation)PIM pass (PIM @firemountain154B ):
docs/tutorials/newcompute)docs/source/modules/documentation)Documentation & Testing Review (@Binglei @przemekff)
Optimizations (To Consider, @przemekff)