Gahow Wang

gahow pushed to main at gahow/kvcache-simulator

2026-04-15 06:49:03 +00:00

365ceac3be chore: update ablation and clean configs

gahow pushed to main at gahow/kvcache-simulator

2026-04-14 07:46:47 +00:00

eaf574cd4e fix: kvcache evict workflow

663ca9c5b9 Support compute_dtype for FP4/FP8 tensor core FLOPS selection

84696604e8 Add B300 GPU preset and GLM-5-NVFP4 on 8xB300 config

Compare 3 commits »

gahow pushed to main at gahow/aituner

2026-04-14 02:27:09 +00:00

bf286ef2a6 docs: add qwen235b prefill 7-day compare

gahow pushed to main at gahow/kvcache-simulator

2026-04-14 02:22:40 +00:00

8d41123418 Update README with full feature documentation

gahow created branch main in gahow/kvcache-simulator

2026-04-13 17:17:05 +00:00

gahow pushed to main at gahow/kvcache-simulator

2026-04-13 17:17:05 +00:00

ec73a95e05 KVCache simulator for LLM serving cluster routing research

gahow created repository gahow/kvcache-simulator

2026-04-13 17:15:04 +00:00

gahow pushed to main at gahow/aituner

2026-04-13 12:50:40 +00:00

26f3b46966 compare: add multi-candidate runner

gahow pushed to main at gahow/aituner

2026-04-13 01:39:34 +00:00

18ff644b32 configs: add qwen235b prefill tight ttft 0323 study

gahow pushed to main at gahow/aituner

2026-04-13 01:37:07 +00:00

bbecec4e9f docs: add qwen235b tight ttft prefill summary

gahow pushed to main at gahow/aituner

2026-04-13 01:33:03 +00:00

ee9ec3c60b docs: add qwen235b decode 0323 summary

gahow pushed to main at gahow/aituner

2026-04-13 01:16:32 +00:00

a1b96f7dd2 docs: update qwen27b 7-day compare

gahow pushed to main at gahow/aituner

2026-04-12 15:09:31 +00:00

4625fba487 trace: make window materialization atomic

gahow pushed to main at gahow/aituner

2026-04-12 14:43:08 +00:00

631a076498 trace: include weekend legacy windows

ade81b5549 docs: add qwen27b chat 0-8k compare summary

Compare 2 commits »

gahow pushed to main at gahow/aituner

2026-04-12 03:24:24 +00:00

edfd61a696 Add qwen235b prefill docs and tight TTFT spec

gahow pushed to main at gahow/aituner

2026-04-11 13:00:04 +00:00

3f20ddf87e Add qwen235b prefill-only tuning support

gahow pushed to main at gahow/aituner

2026-04-11 05:51:55 +00:00

5e54e9c8f5 Add multi-window baseline vs tuned compare flow

gahow pushed to main at gahow/quant

2026-04-11 05:32:44 +00:00

2015b62104 Charge 5 CNY per A-share trade via per-market fee table

gahow pushed to main at gahow/quant

2026-04-11 05:15:10 +00:00

b2176b0c3e Record daily snapshot in cmd_evening for monitor NAV tracking

gahow pushed to main at gahow/aituner

2026-04-11 04:07:44 +00:00

a0b2d7eab2 Add qwen27b and qwen235b tuning notes