Add recipe for HuggingFaceTB-SmolLM2-135M-Instruct (CPU INT4)Add optimization recipe for SmolLM2-135M (CPU INT4)

[Habibur-02]

Description

This PR adds a new optimization recipe for the HuggingFaceTB-SmolLM2-135M-Instruct model targeting CPU execution.

Changes

• Added olive_config.json for INT4 optimization using ModelBuilder.
• Added olive_ci.json for automated testing configuration.
• Added requirements.txt with necessary dependencies.
• Added info.yml for repository indexing.
• Added detailed README.md.

Verification

• Verified locally on Windows CPU environment.
• Model optimized successfully and passed inference tests.

[Copilot]

Pull request overview

This PR adds a new Olive/ONNX Runtime GenAI recipe to optimize the HuggingFaceTB/SmolLM2-135M-Instruct model to INT4 for CPU, including config, CI wiring, metadata, and documentation.

Changes:

• Added olive_config.json defining a CPU INT4 ModelBuilder pipeline for HuggingFaceTB/SmolLM2-135M-Instruct.
• Added olive_ci.json to integrate the new recipe into the GitHub Actions matrix-based CI for Windows CPU.
• Added requirements.txt, info.yml, and README.md to describe, index, and document the recipe.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
HuggingFaceTB-SmolLM2-135M-Instruct/CPU/requirements.txt	Declares Olive, ONNX Runtime, and ONNX Runtime GenAI dependencies needed to run the CPU INT4 recipe.
HuggingFaceTB-SmolLM2-135M-Instruct/CPU/olive_config.json	Defines the Olive configuration for converting and quantizing SmolLM2-135M-Instruct to INT4 on CPU via ModelBuilder.
HuggingFaceTB-SmolLM2-135M-Instruct/CPU/olive_ci.json	Intended to register the Windows CPU INT4 recipe with the CI matrix generator.
HuggingFaceTB-SmolLM2-135M-Instruct/CPU/info.yml	Adds indexing metadata for the SmolLM2 architecture and this CPU INT4 recipe.
HuggingFaceTB-SmolLM2-135M-Instruct/CPU/README.md	Documents how to install dependencies, run the optimization, and perform inference with the generated INT4 model.

---

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

[Habibur-02]

@microsoft-github-policy-service agree