Verify Cezar v3 (1.1893, #2)#90
Closed
willpartcl wants to merge 1 commit into
Closed
Conversation
Verified avg proxy 1.1893 across all 17 IBM benchmarks, 0 overlaps, ~15.5h total runtime. Self-reported 1.037 — verified ~14.7% worse than self-reported, but ~2.7% better than previous Cezar verification (1.2224). Best ibm09=0.9041, worst ibm18=1.4379. Takes #2 above KLA MACH (1.2121). Required two infra changes for the verification: - Added `build-essential` to the eval Dockerfile so torch.compile (Inductor backend) can shell out to gcc to build cached kernels. - Ran with `TORCH_COMPILE_DISABLE=1` and `TORCHDYNAMO_SUPPRESS_ERRORS=1` to bypass a Triton "invalid device context" error during autotuning. Cezar's code has compile fallback at compile time but not at runtime. Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
Contributor
Author
|
Subsumed by #91 which includes the Cezar v3 verified score + 9 new entries + 4 resubmit value updates. |
Mummanajagadeesh
pushed a commit
to Mummanajagadeesh/macro-place-challenge-2026
that referenced
this pull request
May 14, 2026
Cezar v3 verification (subsumes PR partcleda#90): verified 1.1893 (self-reported 1.037). Sits at partcleda#6. Resubmit value updates: - Archgen: 1.3479 → 1.16511 (AutoDMP + GPU, 5/9) - ArzunPD: 1.2478 → 1.1883 (Hyperplace, 5/8) - Binghamton: pending → 1.7621 (feng shui placement, 5/10) - No Man's Sky: note updated (resubmitted 5/6) New entries: - DREAMPlaceProMaxUltra (1.0467, NTHU-NTUST team) — lands partcleda#2 - Vibe (1.1477) — partcleda#3 - Talyxion (1.2075) — partcleda#7 - Adam_A (1.2655) — partcleda#11 - jrslbenn / SPIRAL (1.353) — partcleda#18 - Barsat Khadka / fmpa (1.38) — partcleda#19 - Aegir (1.4553) — partcleda#27 - ZeroLatency (1.5286) — partcleda#35 - rpocevi (1.8894) — partcleda#42 Co-Authored-By: Claude Opus 4.7 (1M context) <[email protected]>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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.
Verified avg proxy 1.1893 across all 17 IBM benchmarks, 0 overlaps, ~15.5h total runtime.
Self-reported 1.037 — verified ~14.7% worse than self-reported, but ~2.7% better than previous Cezar verification (1.2224). Takes #2 above KLA MACH (1.2121).
Required two infra adjustments for the verification:
build-essentialto the eval Dockerfile so torch.compile (Inductor backend) can shell out to gcc to build cached kernels at runtime.TORCH_COMPILE_DISABLE=1andTORCHDYNAMO_SUPPRESS_ERRORS=1to bypass a Triton "invalid device context" error during autotuning. (Cezar's code has compile fallback at compile time but not at runtime.)🤖 Generated with Claude Code