Commit Graph
10 Commits
Author SHA1 Message Date
eSliderandCursor 971852d3af Rework README for better GitHub presentation
Rewrite README with clear value proposition, architecture diagram,
troubleshooting section, and streamlined structure. Update CHANGELOG
to reflect full history of Vulkan-to-SYCL migration.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-12 17:31:33 +00:00
eSliderandCursor c56646e7e7 Switch GPU backend from Vulkan to SYCL for ~2x inference performance on Intel GPUs
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>
2026-02-12 17:28:23 +00:00
eSliderandCursor 63c3b81292 Upgrade ollama from 0.9.3 (IPEX-LLM) to 0.15.6 (official) with Vulkan Intel GPU
Replace the IPEX-LLM portable zip (bundling a patched ollama 0.9.3 with SYCL)
with the official ollama 0.15.6 release using the Vulkan backend for Intel GPU
acceleration. The official ollama project does not ship a SYCL backend; Vulkan
is their supported path for Intel GPUs.

- Use official ollama binary with Vulkan runner (OLLAMA_VULKAN=1)
- Strip CUDA/MLX runners from image to save space
- Add mesa-vulkan-drivers for Intel ANV Vulkan ICD
- Remove all IPEX-LLM env vars and wrapper scripts
- Simplify entrypoint to /usr/bin/ollama serve directly
- Clean up docker-compose.yml: remove IPEX build args and env vars

Tested: Intel Arc Graphics (MTL) detected, 17/17 layers offloaded to Vulkan0
Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-12 15:34:03 +00:00
eSliderandCursor 8debf2010b Fix ollama not reachable from host due to hardcoded OLLAMA_HOST in entrypoint
The IPEX-LLM bundled start-ollama.sh hardcodes OLLAMA_HOST=127.0.0.1 and
OLLAMA_KEEP_ALIVE=10m, overriding docker-compose environment variables and
preventing external connections through Docker port mapping.

- Add custom start-ollama.sh that honours env vars with sensible defaults
- Mount it read-only into the container
- Fix LD_LIBRARY_PATH env var syntax (: -> =)
- Add .gitignore for IDE/swap/webui data files
- Update CHANGELOG and README with fix documentation

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-12 15:18:37 +00:00
eSliderandCursor 96913a2a18 Update Intel GPU stack and ipex-llm to latest available versions
- level-zero v1.22.4 -> v1.28.0
- IGC v2.11.7 -> v2.28.4
- compute-runtime 25.18.33578.6 -> 26.05.37020.3
- libigdgmm 22.7.0 -> 22.9.0
- ipex-llm ollama nightly 2.3.0b20250612 -> 2.3.0b20250725
- Docker compose: disable webui auth, stateless webui volume
- README formatting and GPU model update

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-02-12 15:00:53 +00:00
eSlider 1239010eec Clean up Dockerfile by adding autoremove and autoclean commands to reduce image size 2025-06-21 00:37:38 +01:00
eSlider 0a7f974c04 Update Docker configurations and Intel GPU runtimes for improved performance 2025-06-21 00:35:20 +01:00
eSlider c98fd71fa2 Ignore shelf 2025-04-22 21:15:30 +01:00
eSlider 17592946fa Update Docker configurations for deployment improvements
Revised `IPEXLLM_RELEASE_REPO` value and adjusted file and path references for consistency. Updated `docker-compose.yml` with refined environment variables, device mapping, restart policies, and added necessary port bindings for better functionality and maintainability.
2025-04-22 17:56:04 +01:00
eSliderandGitHub 6a39438666 Fix mess generation bug by updating libraries
* Level Zero 1.18.3
* IGC 1.0.17791
* Compute 24.39.31294.12

https://github.com/mattcurf/ollama-intel-gpu
2024-10-24 15:59:10 +01:00