 Alexander Korolev
|
2d379f9752
final bisect: moe-b12x kept (88, d8192+2.8), linear-b12x is the toxin (44), align perf-negative — recipe locked
|
1 month ago |
|
Alexander Korolev
|
582343c9c6
r17 composition ship anchor: backend-flags are the quality toxin (bisect 46/47/74 vs 87); knobs env clean; graphs+MTP fixed on public r17 refs; ledger
|
1 month ago |
 kilork
|
08a930e477
r12 image: adopt eugr lab build patches (12x patch_vllm_*.py) + pre-staged build sub-deps (SRC_DIR escapes) for flaky node github; ledger: lil render reference + perf push state
|
1 month ago |
|
kilork
|
cc41c48e03
glm53 spark recipe: full lil policy alignment (b12x linear/moe backends, k=5 MTP + moe_backend, instanttensor load, mamba-cache-mode align, INSTANTTENSOR env) + mounted pre-baked JIT caches (~/sm121-cache/glm53-jit-121f, dsv4 pattern); boot 35m -> 7m, exact-391 verified
|
1 month ago |
|
kilork
|
236ce5a7f3
spark recipe: entrypoint (clean image is ENTRYPOINT []) + MAX_JOBS=2 (JIT nvcc OOM at boot); clean r12 image serving, exact-391 verified
|
1 month ago |
|
kilork
|
9395d6d955
r12 clean build: flashinfer-python 0.6.17 from PyPI (whl-index path dead); recipe MTP:3 comment clarified (built-in MTP head, not dspark)
|
1 month ago |
|
kilork
|
8de6340763
clean r12 rebuild: raw-cuda base per solar pattern (r12-specific refs, torch 2.13, flashinfer 0.6.17, b12x PR-head composition); docker_image_dir fixes; bench 88/100 (69-suite)
|
1 month ago |
|
kilork
|
ac5c6c9108
NVFP4-Spark serving on our r12 fork build: spec-config attention_backend=B12X fixes MTP draft selection; intel recipe retargeted; ledger
|
1 month ago |
|
kilork
|
344479943a
recipes: intel glm53 W4A16-AutoRound (tjstewart image, gptq_marlin, entrypoint) + lab NVFP4-Spark (mtp-5 embedded; blocked on whole-network W4A4 loader); ledger: rtx6kpro source contract for own fork build
|
1 month ago |
