Alexander Korolev
|
546e7b2025
glm53-kindling: re-pin BASE to the v0.31.0 release — Docker Hub pruned the entire old nightly-<sha> scheme, the 09-26 base tag is gone and the dir was unbuildable. v0.31.0 (cut 10-02) is the first release containing the base commit; drift audit ddd6fbca..v0.31.0: all patched files unchanged, vllm#58720 landed verbatim upstream (patch dropped, verify-base keeps asserting it), flashinfer 0.7.0->0.7.0.post1, torch/NCCL/cutlass-dsl unchanged
|
1 周之前 |
Alexander Korolev
|
b655965c5f
2x control bench on v2.8.0 (97 scen, hardmode, engine-default effort): 93/100 — the same-harness reference for gr8 (89) and DGPP (88)
|
1 周之前 |
Alexander Korolev
|
989c085a91
chat template: port the merged GLM-5.3 tool-result fix (HF PR #31) — linear sort-merge pairing replaces the O(N^2) id checks; byte-identical output validated on 3000 randomized conversations; 800 parallel calls render 5.7s -> 0.023s (the old path risked vLLM's 30s render timeout). Checksums change: next serve rebuilds both images
|
1 周之前 |
Alexander Korolev
|
0acfa6c680
new-engine quality gates on v2.8.0 (97 scen, hardmode, engine-default effort): gr8 full GLM-5.3 TP=4 = 89/100 (decode 49.9 t/s code c1, prefill 1,294 cold/50.9K cached, agg c4 64.6); DGPP Qwen3.8-Flash-Next NVFP4 w1 = 88/100 (def-effort; the low-effort run 86 on 2.7.1.dev7 is a separate datapoint)
|
1 周之前 |
Alexander Korolev
|
37dc1d74a9
gitignore: per-recipe .lock run ledgers are machine-local state
|
1 周之前 |
Alexander Korolev
|
94ac882a41
glm53-kindling-4x: 30 GiB hot KV pin (measured headroom; pool 5,083,602 tok)
|
1 周之前 |
Alexander Korolev
|
606fe37cb4
glm53-kindling-8x: 48 GiB hot KV pin (measured 24-27 GiB/node headroom at 30G)
|
1 周之前 |
Alexander Korolev
|
58ca8f5034
recipes: glm53-kindling-8x — TP8 on the basic model (b12x load-time head padding, native DFlash2 drafter)
|
1 周之前 |
Alexander Korolev
|
f1c95cde2d
README: cluster config is the add-on's sm121_config_extra option (no second repo)
|
1 周之前 |
Alexander Korolev
|
f24e55b2eb
README: no cluster config in this (public) repo — it lives in the private cluster-config repo
|
1 周之前 |
Alexander Korolev
|
df46bb8b07
docker-images: sparkring-shared
|
1 周之前 |
Alexander Korolev
|
ad8fb377bf
docker-images: solar-open2-nvfp4
|
1 周之前 |
Alexander Korolev
|
24a8341fad
docker-images: qwen38-flashnext-nvfp4
|
1 周之前 |
Alexander Korolev
|
4d470eb805
docker-images: mimo26-spark-vllm
|
1 周之前 |
Alexander Korolev
|
760cb3e782
docker-images: mimo26-spark
|
1 周之前 |
Alexander Korolev
|
50b1f75ee4
docker-images: mia-tp4
|
1 周之前 |
Alexander Korolev
|
5f21f2baec
docker-images: karmic-kraken
|
1 周之前 |
Alexander Korolev
|
bd7380aacd
docker-images: jovian-judgement-r39
|
1 周之前 |
Alexander Korolev
|
eb29c35fdf
docker-images: jovian-judgement
|
1 周之前 |
Alexander Korolev
|
9675392c36
docker-images: glm53-kindling
|
1 周之前 |
Alexander Korolev
|
807beac42b
docker-images: glm53-flash-nvfp4-lab
|
1 周之前 |
Alexander Korolev
|
9ed4603614
docker-images: glm53-flash-nvfp4-4p67
|
1 周之前 |
Alexander Korolev
|
1c07a9588b
docker-images: glm53-flash-nvfp4
|
1 周之前 |
Alexander Korolev
|
b828622aaa
docker-images: glm53-flash-exl3
|
1 周之前 |
Alexander Korolev
|
e2d4af3da4
docker-images: glm53-flash-dflash2
|
1 周之前 |
Alexander Korolev
|
c6eb419752
docker-images: glm53-flash-aiden
|
1 周之前 |
Alexander Korolev
|
3dde0688ae
docker-images: glm52-nf3-hybrid
|
1 周之前 |
Alexander Korolev
|
22b4e641ea
docker-images: glm52
|
1 周之前 |
Alexander Korolev
|
40d2bc57a0
docker-images: dsv41-flash-dcp2
|
1 周之前 |
Alexander Korolev
|
fee7b48cf9
docker-images: dsv41-flash-b12x
|
1 周之前 |