Andriy Oblivantsev
c56646e7e7
Build ggml-sycl from upstream llama.cpp (commit a5bb8ba4, matching ollama's vendored ggml) using Intel oneAPI 2025.1.1 in a multi-stage Docker build. Patch two ollama-specific API divergences via patch-sycl.py: added batch_size parameter to graph_compute, removed GGML_TENSOR_FLAG_COMPUTE skip-check that caused all compute nodes to be bypassed. Tested: gemma3:1b — 27/27 layers on GPU, 10.2 tok/s gen, 65.3 tok/s prompt eval. Co-authored-by: Cursor <cursoragent@cursor.com>
63 lines
1.6 KiB
YAML
63 lines
1.6 KiB
YAML
services:
|
|
ollama-intel-gpu:
|
|
build:
|
|
context: .
|
|
dockerfile: Dockerfile
|
|
args:
|
|
OLLAMA_VERSION: "0.15.6"
|
|
container_name: ollama-intel-gpu
|
|
restart: unless-stopped
|
|
devices:
|
|
- /dev/dri:/dev/dri
|
|
volumes:
|
|
- /tmp/.X11-unix:/tmp/.X11-unix
|
|
- ollama-intel-gpu:/root/.ollama
|
|
shm_size: "16G"
|
|
environment:
|
|
- DISPLAY=${DISPLAY}
|
|
- OLLAMA_HOST=0.0.0.0
|
|
- OLLAMA_DEBUG=1
|
|
- ONEAPI_DEVICE_SELECTOR=level_zero:0
|
|
- ZES_ENABLE_SYSMAN=1
|
|
- OLLAMA_DEFAULT_KEEPALIVE=6h
|
|
- OLLAMA_KEEP_ALIVE=24h
|
|
- OLLAMA_MAX_LOADED_MODELS=1
|
|
- OLLAMA_MAX_QUEUE=512
|
|
- OLLAMA_MAX_VRAM=0
|
|
- OLLAMA_NUM_PARALLEL=1
|
|
#- OLLAMA_NOHISTORY=false
|
|
#- OLLAMA_NOPRUNE=false
|
|
ports:
|
|
- 11434:11434
|
|
|
|
ollama-webui:
|
|
image: ghcr.io/open-webui/open-webui:latest
|
|
container_name: ollama-webui
|
|
volumes:
|
|
- ./webui/data:/app/backend/data
|
|
# - ollama-webui:/app/backend/data
|
|
depends_on:
|
|
- ollama-intel-gpu
|
|
ports:
|
|
- ${OLLAMA_WEBUI_PORT-3000}:8080
|
|
environment:
|
|
- OLLAMA_BASE_URL=http://ollama-intel-gpu:11434
|
|
- OLLAMA_DEFAULT_KEEPALIVE=6h
|
|
#- OPENAI_API_BASE_URL=
|
|
#- OPENAI_API_KEY=
|
|
#
|
|
# AUTOMATIC1111_BASE_URL="http://localhost:7860"
|
|
- WEBUI_AUTH=False
|
|
- ENABLE_RAG_WEB_SEARCH=True
|
|
|
|
# DO NOT TRACK
|
|
- SCARF_NO_ANALYTICS=true
|
|
- DO_NOT_TRACK=true
|
|
- ANONYMIZED_TELEMETRY=false
|
|
extra_hosts:
|
|
- host.docker.internal:host-gateway
|
|
restart: unless-stopped
|
|
volumes:
|
|
ollama-webui: {}
|
|
ollama-intel-gpu: {}
|