From 389ddbc3148209bae4c6ac3429b53edcbf2f5690 Mon Sep 17 00:00:00 2001 From: Marcel Klehr Date: Wed, 31 Jul 2024 13:39:06 +0200 Subject: [PATCH] docs(AI): Add more details to requirements Signed-off-by: Marcel Klehr --- admin_manual/ai/app_context_chat.rst | 4 ++++ admin_manual/ai/app_stt_whisper2.rst | 1 + admin_manual/ai/app_translate.rst | 4 +++- 3 files changed, 8 insertions(+), 1 deletion(-) diff --git a/admin_manual/ai/app_context_chat.rst b/admin_manual/ai/app_context_chat.rst index 9fc97f2cf..b80a2a823 100644 --- a/admin_manual/ai/app_context_chat.rst +++ b/admin_manual/ai/app_context_chat.rst @@ -31,6 +31,10 @@ Requirements * CPU Sizing * At least 12GB of system RAM + * Below version 3, 10-20 Cores, the more cores the faster it will be + * Since version 3, this app makes use of the configured Text To Text Free prompt provider instead of running its own Language model, you will thus need only 4-8 cores for the embedding model + +* A dedicated machine is recommended Space usage ~~~~~~~~~~~ diff --git a/admin_manual/ai/app_stt_whisper2.rst b/admin_manual/ai/app_stt_whisper2.rst index d2444facf..9850e609d 100644 --- a/admin_manual/ai/app_stt_whisper2.rst +++ b/admin_manual/ai/app_stt_whisper2.rst @@ -25,6 +25,7 @@ Requirements * The more cores you have and the more powerful the CPU the better, we recommend 10-20 cores * The app will hog all cores by default, so it is usually better to run it on a separate machine + * 4GB for the app Installation ------------ diff --git a/admin_manual/ai/app_translate.rst b/admin_manual/ai/app_translate.rst index e4ea8b265..3ebcf8e92 100644 --- a/admin_manual/ai/app_translate.rst +++ b/admin_manual/ai/app_translate.rst @@ -20,9 +20,11 @@ Requirements ------------ * Minimal Nextcloud version: 26 -* x86 CPU +* x86 CPU with 4-8 cores for the app to use (The more cores the faster it will be) +* 2GB of RAM for the app should be enough * GNU lib C (musl is not supported) * This app does not support using GPU for processing and may thus not be performing ideally for long texts +* The workload will run on the web server workers (*Note*: Nextcloud AIO is currently not supported due to it using musl)