LocalAI

mirror of https://github.com/mudler/LocalAI.git synced 2026-06-23 16:19:07 -04:00

Files

Ettore Di Giacinto faeb5b457c analysis: NVFP4 closes the decode gap too (547->619, ~93% of vLLM)

Measured npl=128 cold A/B: NVFP4 decode 619 vs Q4_K 547 (+13%), closing the gap to
vLLM (667) from ~22% to ~7%. NVFP4's FP4-MMA kernel is more bandwidth-efficient at
the thin n=128 decode shape than Q4_K int8-MMQ (which ran 2.1x above the floor), so
it IS the better int4 decode GEMM the diagnosis called for - no multi-day
Marlin-for-K-quants needed. With NVFP4, llama.cpp on GB10 is ahead on prefill
(1209 vs 800) and within ~7% on decode. Remaining 7% = optional FP4 kernel tuning.

Assisted-by: Claude:opus-4.8 [Claude Code]
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

2026-06-21 21:42:17 +00:00

paged

analysis: NVFP4 closes the decode gap too (547->619, ~93% of vLLM)

2026-06-21 21:42:17 +00:00

patches

bench(dense): root-cause the W4A4 NVFP4 hang; W4A16 vs Q4 is the headline

2026-06-20 06:59:50 +00:00

CMakeLists.txt

fix(turboquant): resolve common.h by detecting llama-common vs common target (#9413 )

2026-04-18 20:30:28 +02:00

grpc-server.cpp

feat: generic chat_template_kwargs (model config + per-request metadata) (#10359 )