Sync with upstream ollama/ollama and restore Tesla K80 (compute 3.7) support
This commit represents a complete rework after pulling the latest changes from official ollama/ollama repository and re-applying Tesla K80 compatibility patches. ## Key Changes ### CUDA Compute Capability 3.7 Support (Tesla K80) - Added sm_37 (compute 3.7) to CMAKE_CUDA_ARCHITECTURES in CMakeLists.txt - Updated CMakePresets.json to include compute 3.7 in "CUDA 11" preset - Using 37-virtual (PTX with JIT compilation) for maximum compatibility ### Legacy Toolchain Compatibility - **NVIDIA Driver**: 470.256.02 (last version supporting Kepler/K80) - **CUDA Version**: 11.4.4 (last CUDA 11.x supporting compute 3.7) - **GCC Version**: 10.5.0 (required by CUDA 11.4 host_config.h) ### CPU Architecture Trade-offs Due to GCC 10.5 limitation, sacrificed newer CPU optimizations: - Alderlake CPU variant enabled WITHOUT AVX_VNNI (requires GCC 11+) - Still supports: SSE4.2, AVX, F16C, AVX2, BMI2, FMA - Performance impact: ~3-7% on newer CPUs (acceptable for K80 compatibility) ### Build System Updates - Modified ml/backend/ggml/ggml/src/ggml-cuda/CMakeLists.txt for compute 3.7 - Added -Wno-deprecated-gpu-targets flag to suppress warnings - Updated ml/backend/ggml/ggml/src/CMakeLists.txt for Alderlake without AVX_VNNI ### Upstream Sync Merged latest llama.cpp changes including: - Enhanced KV cache management with ISWA and hybrid memory support - Improved multi-modal support (mtmd framework) - New model architectures (Gemma3, Llama4, Qwen3, etc.) - GPU backend improvements for CUDA, Metal, and ROCm - Updated quantization support and GGUF format handling ### Documentation - Updated CLAUDE.md with comprehensive build instructions - Documented toolchain constraints and CPU architecture trade-offs - Removed outdated CI/CD workflows (tesla-k80-*.yml) - Cleaned up temporary development artifacts ## Rationale This fork maintains Tesla K80 GPU support (compute 3.7) which was dropped in official Ollama due to legacy driver/CUDA requirements. The toolchain constraint creates a deadlock: - K80 → Driver 470 → CUDA 11.4 → GCC 10 → No AVX_VNNI We accept the loss of cutting-edge CPU optimizations to enable running modern LLMs on legacy but still capable Tesla K80 hardware (12GB VRAM per GPU). 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
BIN
docs/images/cline-mcp.png
Normal file
|
After Width: | Height: | Size: 556 KiB |
BIN
docs/images/cline-settings.png
Normal file
|
After Width: | Height: | Size: 76 KiB |
BIN
docs/images/codex-mcp.png
Normal file
|
After Width: | Height: | Size: 948 KiB |
BIN
docs/images/favicon.png
Normal file
|
After Width: | Height: | Size: 890 B |
BIN
docs/images/goose-cli.png
Normal file
|
After Width: | Height: | Size: 160 KiB |
BIN
docs/images/goose-mcp-1.png
Normal file
|
After Width: | Height: | Size: 877 KiB |
BIN
docs/images/goose-mcp-2.png
Normal file
|
After Width: | Height: | Size: 911 KiB |
BIN
docs/images/goose-settings.png
Normal file
|
After Width: | Height: | Size: 109 KiB |
BIN
docs/images/intellij-chat-sidebar.png
Normal file
|
After Width: | Height: | Size: 69 KiB |
BIN
docs/images/intellij-current-model.png
Normal file
|
After Width: | Height: | Size: 106 KiB |
BIN
docs/images/intellij-local-models.png
Normal file
|
After Width: | Height: | Size: 79 KiB |
BIN
docs/images/logo-dark.png
Normal file
|
After Width: | Height: | Size: 3.3 KiB |
BIN
docs/images/logo.png
Normal file
|
After Width: | Height: | Size: 2.7 KiB |
BIN
docs/images/n8n-chat-model.png
Normal file
|
After Width: | Height: | Size: 87 KiB |
BIN
docs/images/n8n-chat-node.png
Normal file
|
After Width: | Height: | Size: 70 KiB |
BIN
docs/images/n8n-credential-creation.png
Normal file
|
After Width: | Height: | Size: 43 KiB |
BIN
docs/images/n8n-models.png
Normal file
|
After Width: | Height: | Size: 130 KiB |
BIN
docs/images/n8n-ollama-form.png
Normal file
|
After Width: | Height: | Size: 53 KiB |
BIN
docs/images/ollama-settings.png
Normal file
|
After Width: | Height: | Size: 3.6 MiB |
BIN
docs/images/vscode-model-options.png
Normal file
|
After Width: | Height: | Size: 77 KiB |
BIN
docs/images/vscode-models.png
Normal file
|
After Width: | Height: | Size: 56 KiB |
BIN
docs/images/vscode-sidebar.png
Normal file
|
After Width: | Height: | Size: 25 KiB |
BIN
docs/images/welcome.png
Normal file
|
After Width: | Height: | Size: 233 KiB |
BIN
docs/images/xcode-chat-icon.png
Normal file
|
After Width: | Height: | Size: 186 KiB |
BIN
docs/images/xcode-intelligence-window.png
Normal file
|
After Width: | Height: | Size: 182 KiB |
BIN
docs/images/xcode-locally-hosted.png
Normal file
|
After Width: | Height: | Size: 146 KiB |
BIN
docs/images/zed-ollama-dropdown.png
Normal file
|
After Width: | Height: | Size: 38 KiB |
BIN
docs/images/zed-settings.png
Normal file
|
After Width: | Height: | Size: 57 KiB |