[II] Serve projection-mixed EXL3 K3/K4/K5 weights - #148
voipmonitor wants to merge 2 commits into
Conversation
Implement one graph-capturable mixed-Trellis launch plan for EXL3 checkpoints that assign MCG K3, K4, and K5 independently to gate, up, and down projections. Validate descriptor counts and packed extents before launch, retain graph-stable scratch ownership, and prewarm route, rank-slice, and scalar-alignment specializations. Preserve the current QSRT atoms-v2 pair contract and the removal of the route-major W4A8 interface.
|
Important Review skippedDraft detected. Please check the settings in the CodeRabbit UI or the ⚙️ Run configurationConfiguration used: Organization UI Review profile: CHILL Plan: Pro Plus Run ID: You can disable this status message by setting the Use the checkbox below for a quick retry:
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
Closing without merge because this branch is incompatible with the The host-only route-pack specialization prewarm idea is separable and should be proposed independently from |
Status
Purpose
Serve EXL3 checkpoints that assign MCG K3, K4, and K5 independently to routed-expert gate, up, and down projections. The B12X runtime reads packed checkpoint storage directly and executes all projection tiers through one CUDA-graph-capturable mixed-Trellis launch plan.
Resulting behavior
Compatibility
The MCG codebook accepts K3, K4, and K5 for this mixed runtime. SQG-XOR-Cheb-T12 retains its K2/K3/K4 contract. Existing QSRT paths and the removed route-major W4A8 interface are unchanged. Malformed or incomplete projection maps fail closed.
The vLLM checkpoint loader is provided by a companion pull request against
dev/infernal-invocation.Validation
Source base:
ac07f6746edce4852f01cc11f34dbe7abae1fdaa.git diff --check: passed.