Ettore Di Giacinto a64fc865ba fix(distributed): aggregate prefix cache pressure ...

Broadcast forced-disturb events and successful scale-up resets so every frontend shares the same rolling autoscale signal. Deduplicate NATS echoes, expose an origin-only Prometheus counter, and document cluster behavior.

Closes #10083

Assisted-by: Codex:gpt-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>

2026-09-11 21:48:07 +00:00

..

audio

feat(localvqe/audio): v1.3 release and add spectrograms to audio transform UI (#10113)

2026-05-31 23:56:46 +02:00

clusterrouting

refactor(routing): extract replica picker into pkg/clusterrouting (#10123)

2026-06-01 09:38:55 +02:00

concurrency

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

distributedhdr

feat: prefix-cache-aware routing for distributed mode (#10071)

2026-05-30 23:24:22 +02:00

downloader

fix(downloader): make file:// installs reachable again (#11701) (#11734)

2026-09-03 18:30:31 +02:00

functions

fix(grammars): restore backslash escaping in llama31 grammar fixture (#11242)

2026-07-30 23:53:23 +02:00

grpc

feat(router): make KNN a first-class classifier with a persisted, curated corpus (#10652)

2026-08-18 09:37:43 +02:00

httpclient

Revert "chore(tests): Avoid network, sleep and more during tests" (#11601)

2026-08-19 16:39:39 +02:00

huggingface-api

Revert "chore(tests): Avoid network, sleep and more during tests" (#11601)

2026-08-19 16:39:39 +02:00

mcp/localaitools

feat(distributed): key scheduling rules by a model alias (#11771)

2026-08-29 09:16:50 +02:00

model

[model-config] feat: add environment variables support for backends (#10721)

2026-09-11 18:02:08 +02:00

modelartifacts

Revert "chore(tests): Avoid network, sleep and more during tests" (#11601)

2026-08-19 16:39:39 +02:00

natsauth

fix(distributed): aggregate prefix cache pressure

2026-09-11 21:48:07 +00:00

oci

fix(oci): resume interrupted layer downloads (#11688)

2026-08-24 09:33:44 +02:00

radixtree

feat: prefix-cache-aware routing for distributed mode (#10071)

2026-05-30 23:24:22 +02:00

reasoning

fix(reasoning): stop prefilled <think> from swallowing tag-less answers (#10225)

2026-06-09 09:02:04 +02:00

safefile

fix(distributed): bound ephemeral staging (#11924)

2026-09-08 18:13:40 +02:00

sanitize

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

signals

feat: add distributed mode (#9124)

2026-03-30 00:47:27 +02:00

sound

feat(realtime): Semantic VAD EOU token (#10444)

2026-06-30 09:01:22 +02:00

store

fix(backends): refuse foreign model loads in opus and local-store (#10769)

2026-07-11 09:17:34 +02:00

system

feat(gallery): let one gallery entry offer several builds of the same model (#10943)

2026-07-20 18:43:02 +02:00

tokens

feat(chat): add end-to-end context compression (#11556)

2026-08-18 11:31:03 +00:00

utils

fix(utils): reject tar hardlinks that escape the extraction root (#11266)

2026-08-01 09:25:35 +02:00

vram

fix(vram): persist remote probe metadata (#11487)

2026-08-23 08:55:13 +02:00

vrambudget

feat(vram): per-node VRAM allocation budget (LOCALAI_VRAM_BUDGET) (#10833)

2026-07-15 09:58:45 +02:00

xio

feat(xio): make copy buffer size configurable (#11660)

2026-08-21 16:14:26 +02:00

xsync

chore: fix go.mod module (#2635)

2024-06-23 08:24:36 +00:00

xsysinfo

fix(nodes): never schedule a model onto a node that cannot store it (#11054)

2026-07-23 00:03:21 +02:00