mirror of
https://github.com/getumbrel/umbrel-apps.git
synced 2026-11-20 21:23:41 +03:00
Update localai to v5.0.0
Co-authored-by: Kai Chappel <kaichappel@users.noreply.github.com> Co-authored-by: nmfretz <85373263+nmfretz@users.noreply.github.com>
This commit is contained in:
2 files changed
+12
-9
No files matched your search
@@ -7,7 +7,8 @@ services:
|
||||
APP_PORT: 8080
|
||||
PROXY_AUTH_WHITELIST: "/v1/*"
|
||||
api:
|
||||
image: localai/localai:v4.12.1@sha256:c8556c4f7b6dd9b76e915d515ab5cfe5a765d8ea8fd3f0cf563f9c690b48a7f4
|
||||
image: localai/localai:v5.0.0@sha256:4ef0d723ccb18cdd653cf7416d548cb6add42d1fcc6459d655d4c9f803d03dea
|
||||
restart: on-failure
|
||||
healthcheck:
|
||||
test: ["CMD", "curl", "-f", "http://localhost:8080/readyz"]
|
||||
interval: 1m
|
||||
@@ -17,6 +18,8 @@ services:
|
||||
- MODELS_PATH=/models
|
||||
- LOCALAI_BACKENDS_PATH=/backends
|
||||
- LOCALAI_EXTERNAL_BACKENDS=cpu-llama-cpp
|
||||
# Honor the external HTTPS scheme forwarded by the Umbrel app gateway.
|
||||
- LOCALAI_TRUSTED_PROXIES=${GATEWAY_IP}
|
||||
# The image pins its backend capability to CPU; point the override file at a missing path so
|
||||
# LocalAI detects the host GPU and offers matching CUDA/ROCm/SYCL backends in its gallery
|
||||
- LOCALAI_FORCE_META_BACKEND_CAPABILITY_RUN_FILE=/nonexistent
|
||||
|
||||
@@ -4,7 +4,7 @@ storage:
|
||||
dataRoot: data
|
||||
category: ai
|
||||
name: LocalAI
|
||||
version: "v4.12.1-patch.1"
|
||||
version: "v5.0.0"
|
||||
tagline: Drop-in OpenAI replacement
|
||||
description: >-
|
||||
LocalAI is the free, Open Source OpenAI alternative. LocalAI act as a drop-in replacement REST API that's compatible with OpenAI API specifications for local inferencing.
|
||||
@@ -18,17 +18,17 @@ description: >-
|
||||
Before running a model, make sure your device has enough free RAM to support it. Attempting to run a model that exceeds your available memory could cause your device to crash or become unresponsive. Always check the model requirements before downloading or starting it.
|
||||
permissions:
|
||||
- GPU
|
||||
releaseNotes: >-
|
||||
This update adds support for new storage management and configuration options introduced with umbrelOS 2.0.
|
||||
releaseNotes: |-
|
||||
LocalAI 5.0 adds server-side chats, projects, file management, and a refreshed interface.
|
||||
|
||||
Chats are now stored on the server. Without separate LocalAI accounts, people with access to this app share the same chat history. To keep older browser-only chats, follow the upgrade guide to export and import them from the browser that holds them.
|
||||
|
||||
Previous LocalAI v4.12.1 release notes:
|
||||
New chats turn on agent tools by default, which need more context than small models have. If a chat fails with a context window error, turn tools off in the chat or use a model with a larger context size.
|
||||
|
||||
Studio and Talk are now available as separate add-ons. Models configured with the removed liquid-audio backend need a compatible replacement.
|
||||
|
||||
Security update that strengthens protections when fetching user-supplied image, audio, and video URLs. Redirect destinations are now validated at every step, and connections use the validated address to prevent access to private or reserved networks.
|
||||
|
||||
|
||||
Full release notes can be found at https://github.com/mudler/LocalAI/releases/tag/v4.12.1
|
||||
Upgrade guide: https://localai.io/docs/features/chats-migration/
|
||||
Full release notes: https://github.com/mudler/LocalAI/releases/tag/v5.0.0
|
||||
backupIgnore:
|
||||
- data/models/*
|
||||
- data/backends/*
|
||||
|
||||
Reference in new issue
Block a user