From a0b44b2ca204d2d7d6a817d2a73f694acf698507 Mon Sep 17 00:00:00 2001 From: Michael Mayer Date: Mon, 1 Sep 2025 16:03:27 +0200 Subject: [PATCH] Setup: Update Ollama service examples in compose.yaml files #5123 Signed-off-by: Michael Mayer --- compose.nvidia.yaml | 2 +- compose.yaml | 2 +- setup/docker/arm64/compose.yaml | 2 +- setup/docker/compose.yaml | 2 +- setup/docker/nvidia/compose.yaml | 2 +- 5 files changed, 5 insertions(+), 5 deletions(-) diff --git a/compose.nvidia.yaml b/compose.nvidia.yaml index b5d476da6..8225ea058 100644 --- a/compose.nvidia.yaml +++ b/compose.nvidia.yaml @@ -179,7 +179,7 @@ services: OLLAMA_NOPRUNE: "false" # disables pruning of model blobs at startup OLLAMA_NOHISTORY: "true" # disables readline history OLLAMA_FLASH_ATTENTION: "false" # enables the experimental flash attention feature - OLLAMA_KV_CACHE_TYPE: "f16" # see https://mitjamartini.com/blog/kv-cache-quantization-in-ollama/ + OLLAMA_KV_CACHE_TYPE: "f16" # cache quantization (f16, q8_0, or q4_0) OLLAMA_SCHED_SPREAD: "false" # allows scheduling models across all GPUs. OLLAMA_NEW_ENGINE: "true" # enables the new Ollama engine # OLLAMA_DEBUG: "true" # shows additional debug information diff --git a/compose.yaml b/compose.yaml index 568434a70..3187c046c 100644 --- a/compose.yaml +++ b/compose.yaml @@ -259,7 +259,7 @@ services: OLLAMA_NOPRUNE: "false" # disables pruning of model blobs at startup OLLAMA_NOHISTORY: "true" # disables readline history OLLAMA_FLASH_ATTENTION: "false" # enables the experimental flash attention feature - OLLAMA_KV_CACHE_TYPE: "f16" # see https://mitjamartini.com/blog/kv-cache-quantization-in-ollama/ + OLLAMA_KV_CACHE_TYPE: "f16" # cache quantization (f16, q8_0, or q4_0) OLLAMA_SCHED_SPREAD: "false" # allows scheduling models across all GPUs. OLLAMA_NEW_ENGINE: "true" # enables the new Ollama engine # OLLAMA_DEBUG: "true" # shows additional debug information diff --git a/setup/docker/arm64/compose.yaml b/setup/docker/arm64/compose.yaml index 226d834f4..0ea068eb6 100644 --- a/setup/docker/arm64/compose.yaml +++ b/setup/docker/arm64/compose.yaml @@ -181,7 +181,7 @@ services: OLLAMA_NOPRUNE: "false" # disables pruning of model blobs at startup OLLAMA_NOHISTORY: "true" # disables readline history OLLAMA_FLASH_ATTENTION: "false" # enables the experimental flash attention feature - OLLAMA_KV_CACHE_TYPE: "f16" # see https://mitjamartini.com/blog/kv-cache-quantization-in-ollama/ + OLLAMA_KV_CACHE_TYPE: "f16" # cache quantization (f16, q8_0, or q4_0) OLLAMA_SCHED_SPREAD: "false" # allows scheduling models across all GPUs. OLLAMA_NEW_ENGINE: "true" # enables the new Ollama engine # OLLAMA_DEBUG: "true" # shows additional debug information diff --git a/setup/docker/compose.yaml b/setup/docker/compose.yaml index c99de6224..d43219336 100644 --- a/setup/docker/compose.yaml +++ b/setup/docker/compose.yaml @@ -186,7 +186,7 @@ services: OLLAMA_NOPRUNE: "false" # disables pruning of model blobs at startup OLLAMA_NOHISTORY: "true" # disables readline history OLLAMA_FLASH_ATTENTION: "false" # enables the experimental flash attention feature - OLLAMA_KV_CACHE_TYPE: "f16" # see https://mitjamartini.com/blog/kv-cache-quantization-in-ollama/ + OLLAMA_KV_CACHE_TYPE: "f16" # cache quantization (f16, q8_0, or q4_0) OLLAMA_SCHED_SPREAD: "false" # allows scheduling models across all GPUs. OLLAMA_NEW_ENGINE: "true" # enables the new Ollama engine # OLLAMA_DEBUG: "true" # shows additional debug information diff --git a/setup/docker/nvidia/compose.yaml b/setup/docker/nvidia/compose.yaml index 2d2538e3b..777902145 100644 --- a/setup/docker/nvidia/compose.yaml +++ b/setup/docker/nvidia/compose.yaml @@ -186,7 +186,7 @@ services: OLLAMA_NOPRUNE: "false" # disables pruning of model blobs at startup OLLAMA_NOHISTORY: "true" # disables readline history OLLAMA_FLASH_ATTENTION: "false" # enables the experimental flash attention feature - OLLAMA_KV_CACHE_TYPE: "f16" # see https://mitjamartini.com/blog/kv-cache-quantization-in-ollama/ + OLLAMA_KV_CACHE_TYPE: "f16" # cache quantization (f16, q8_0, or q4_0) OLLAMA_SCHED_SPREAD: "false" # allows scheduling models across all GPUs. OLLAMA_NEW_ENGINE: "true" # enables the new Ollama engine # OLLAMA_DEBUG: "true" # shows additional debug information