Alexander Korolev
|
9c648a2ee4
state doc: night-3 close — final campaign table, DFlash2 blocker, queue
|
1 개월 전 |
Alexander Korolev
|
9cb56de106
state doc: night 2 results (aiden 159 first canonical, r27 156), next queue
|
1 개월 전 |
Alexander Korolev
|
3e63d19947
state doc: image build v2 (clone-at-pin) contract
|
1 개월 전 |
Alexander Korolev
|
0a1a5dbcc1
typo
|
1 개월 전 |
Alexander Korolev
|
b564bd561d
ops layout: 2nodes-dac = production (operator agent), ds4 = dev lane
|
1 개월 전 |
Alexander Korolev
|
ca8bb6f448
campaign extended: 2x nvfp4 KV (154), 2x Spark (158, best 2x), 4x Spark blocked (mm_mxfp8 TP4); 4x 16G pins; JIT warm-start 9min
|
1 개월 전 |
Alexander Korolev
|
cc19b5f581
persist JIT caches per node (flashinfer/torchinductor/vllm mounts in all glm53 lanes)
|
1 개월 전 |
Alexander Korolev
|
24cfdc3a76
state doc: add --perf comparison table (all 7 lanes, llama-benchy pp/tg/TTFT)
|
1 개월 전 |
Alexander Korolev
|
5d587215dd
separation complete: r18 dir carries source tarballs; state doc records r12/r18 split + r17 rebuild caveat
|
1 개월 전 |
Alexander Korolev
|
fe070bebdd
night shift: r18 built + full lane matrix (DCP4 161/176 best; eugr/DFlash2/dsv4fv-eugr documented)
|
1 개월 전 |
Alexander Korolev
|
ccd615fcad
fix hardmode for night shift + future: version-pin tool (2.6.1.dev45, --hardmode = 88 scenarios), re-baseline anchor lane before gating; reports commit to recipes/<lane>/reports/; new eugr-4x test recipe (2x stays production default until 4x dramatic)
|
1 개월 전 |
Alexander Korolev
|
54dfffeb24
eugr GLM lane: document forum 381350#364 parity + DFlash2 alt spec (night queue 3b: k=7 incoai drafter A/B vs MTP k=5)
|
1 개월 전 |
Alexander Korolev
|
82a8368e5b
night queue: DCP4 probe (sparkring reference, r17abi-first, no build dep) + #646 geometry check; refresh r18 pins to 343661fe/1adb8a38
|
1 개월 전 |
Alexander Korolev
|
f8ab660d6c
new lane dsv4-flash-vision-eugr (eugr b12x nightly + Vision-Exp, DSpark-3 adaptive, plain-B12X attn); night-shift queue: r18 build, redhat MTP, eugr GLM, dsv4-vision eugr
|
1 개월 전 |
Alexander Korolev
|
f1d9e635f0
new lane glm53-flash-nvfp4-eugr: eugr B12X full-cluster dual-Spark variant (nightly-20260904 image, RoCE one-shot, MTP k=5 humming) — A/B cell vs r18; dry-run verified
|
1 개월 전 |
Alexander Korolev
|
02b94db96f
redhat lane: add MTP k=5 — model_mtp.safetensors (7.6 GB) now in the main NVFP4 repo; card 'Enable Speculative Decoding' section (2026-09-05)
|
1 개월 전 |
Alexander Korolev
|
509654fbed
ledger: glm53-flash-aiden-4x row (89/156-176, 37.6 tg c1 flat to d8192, prefill 2,185 = his ref) + TC-43 note
|
1 개월 전 |
Alexander Korolev
|
6922cb8584
ledger: glm53-flash-aiden-2x row (aiden production-1.1 + incoai DFlash2 k=7, 89/157-176)
|
1 개월 전 |
Alexander Korolev
|
fd4c74dcc1
checkpoint A/B on ds4: NVFP4 lab ckpt = same 88 quality on r17 stack but +4-7 tok/s; lab day-0 stack retains 93; r17abi recipe added
|
1 개월 전 |
Alexander Korolev
|
04984a169d
ledger: one-backend bisect solved (linear b12x = the toxin); ship config locked
|
1 개월 전 |
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 개월 전 |
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 개월 전 |
kilork
|
5283d11ce1
clean r12 image shipped: 90/100 c1 (was 88 day-0-base), entrypoint+MAX_JOBS boot fixes, ledger
|
1 개월 전 |
kilork
|
2d17cd0356
r12 clean build: SHA-pinned PR heads (550 closed), deepgemm pre-clone, lil launch-policy diff recorded
|
1 개월 전 |
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 개월 전 |
kilork
|
ac5c6c9108
NVFP4-Spark serving on our r12 fork build: spec-config attention_backend=B12X fixes MTP draft selection; intel recipe retargeted; ledger
|
1 개월 전 |
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 개월 전 |
kilork
|
de62327a2b
ledger: dsv4vision tool-eval-bench c1 (81/100, tg 34.1/31.0/29.1, deployability 74, 3 safety-critical fails)
|
1 개월 전 |
kilork
|
8d49ca7159
dsv4vision: env-gated openai mm branch (VLLM_DSV4_MM), worker shape-follow patches, local-NVMe weights (dgx3+dgx4), cache-clear runbook; text lane verified exact on new boots
|
1 개월 전 |
kilork
|
d34e7758bc
dsv4-vision lane: text serving verified on eugr fork (34.9 tok/s, accept 0.714); OpenAI-mm processor gap documented
|
1 개월 전 |