Skip to content

updates to containerization - #1

Merged
beatgeek merged 3 commits into
mainfrom
gpu-support
Jan 18, 2026
Merged

updates to containerization#1
beatgeek merged 3 commits into
mainfrom
gpu-support

Conversation

@beatgeek

Copy link
Copy Markdown
Owner

Containerization!

@beatgeek
beatgeek requested a review from Copilot January 18, 2026 21:41
@beatgeek beatgeek self-assigned this Jan 18, 2026

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR adds containerization support for KernelPort with separate CPU and GPU Docker images, updates the gRPC API to use proper protobuf enums for data types, and implements a model fetching mechanism for HuggingFace.

Changes:

  • Added Docker support with CPU (Dockerfile.cpu) and GPU (Dockerfile.gpu) images, including a model fetch script and entrypoint
  • Updated gRPC API from Predict to Infer with proper DType enum instead of string-based types
  • Added documentation for model ingestion, validation, and manifest schema

Reviewed changes

Copilot reviewed 24 out of 26 changed files in this pull request and generated no comments.

Show a summary per file
File Description
scripts/model_fetch.py Model download script for HuggingFace repositories
scripts/entrypoint.sh Container entrypoint that optionally fetches models before starting server
models/sample/README.md Documentation for sample model bundle
models/sample-manifest.yaml Example manifest for HuggingFace model configuration
docs/validation.md Validation instructions for CPU/GPU containers and gRPC testing
docs/model-manifest.md Proposed schema for model manifest files
docs/model-ingestion.md Model ingestion strategy documentation
crates/kernelport-server/src/registry.rs Extracts and stores output names from loaded models
crates/kernelport-server/src/main.rs Updates to use model path CLI arg, adds reflection, improves error handling
crates/kernelport-server/src/grpc.rs Migrates from string-based to enum-based DType handling
crates/kernelport-server/src/cli.rs Adds model-path CLI argument
crates/kernelport-server/Cargo.toml Upgrades tonic to 0.14 and adds tonic-reflection
crates/kernelport-proto/src/lib.rs Exports file descriptor set for gRPC reflection
crates/kernelport-proto/src/inference.proto Adds DType enum and renames Predict to Infer
crates/kernelport-proto/build.rs Generates file descriptor set for reflection
crates/kernelport-proto/Cargo.toml Upgrades to tonic/prost 0.14
crates/kernelport-backend-ort/tests/identity.rs Code formatting improvements
crates/kernelport-backend-ort/src/lib.rs Code formatting and clippy fixes
README.md Adds container documentation and pre-commit instructions
Dockerfile.gpu GPU-enabled container with CUDA runtime
Dockerfile.cpu CPU-only container with Debian runtime
.pre-commit-config.yaml Pre-commit configuration for cargo fmt
.github/copilot-instructions.md Copilot guidance for repository conventions
.dockerignore Docker build exclusions

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@beatgeek
beatgeek requested a review from Copilot January 18, 2026 21:46

Copilot AI left a comment

Copy link
Copy Markdown

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

Copilot reviewed 25 out of 27 changed files in this pull request and generated no new comments.


💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

@beatgeek
beatgeek merged commit baa9ca7 into main Jan 18, 2026
2 checks passed
@beatgeek
beatgeek deleted the gpu-support branch January 18, 2026 21:50
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants