Make ONNX graph optimization configurable per model (default off)#37252
Merged
Conversation
Previously the ONNX graph optimization level was hardcoded to ENABLE (ORT_ENABLE_ALL) both in the proton model cache and the container stateless evaluator. Add an "optimize-model" setting to the onnx-model schema block, wired through onnx-models.def, that controls this per model. Default is false (DISABLE / NO_OPT), so optimization is off unless explicitly enabled. Proton ranking path: - onnx-models.def: new model[].optimize_model bool default=false - OnnxModelCache::load takes an Onnx::Optimize argument and includes it in the cache key, so the same file with different optimize settings no longer shares an instance - fef::OnnxModel carries the bool; onnx_feature translates it to Onnx::Optimize when loading from the cache Container stateless path: - OnnxEvaluatorOptions gains an optimizeModel field (part of the session cache key hash), honored in EmbeddedOnnxRuntime.createSessionOptions (ALL_OPT vs NO_OPT) - RankProfilesConfigImporter reads optimize_model() from config Parsers: - SchemaParser.jj and the schema-language-server SchemaParser.ccc both accept "optimize-model: <bool>" in an onnx-model block Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
havardpe
approved these changes
Jun 19, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Previously the ONNX graph optimization level was hardcoded to ENABLE (ORT_ENABLE_ALL) both in the proton model cache and the container stateless evaluator. Add an "optimize-model" setting to the onnx-model schema block, wired through onnx-models.def, that controls this per model. Default is false (DISABLE / NO_OPT), so optimization is off unless explicitly enabled.
Proton ranking path:
Container stateless path:
Parsers: