Bfcl v4 #3905
Open
Bfcl v4 #3905
+324
−105
Conversation
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
atobiszei
reviewed
Jan 13, 2026
tests/accuracy/test_small_models.sh
Outdated
| @@ -0,0 +1,86 @@ | |||
| #!/bin/bash -x | |||
| # | |||
| # Copyright (c) 2024 Intel Corporation | |||
| @@ -0,0 +1,22 @@ | |||
| export MODEL=$1 | |||
Contributor
There was a problem hiding this comment.
Pull request overview
This PR updates the Berkeley Function Call Leaderboard (BFCL) integration to version 4, refactoring test infrastructure and adding support for new models and chat templates.
Changes:
- Refactored testing scripts by extracting model test logic from
export_all_models.shinto dedicated test scripts - Updated gorilla benchmark integration to a newer version with modified patch configurations
- Added support for additional models (GPT-OSS-20B, Qwen3-Coder-30B, Devstral) with corresponding chat templates
Reviewed changes
Copilot reviewed 8 out of 8 changed files in this pull request and generated 4 comments.
Show a summary per file
| File | Description |
|---|---|
| tests/accuracy/test_small_models.sh | New script containing extracted model testing logic with configurable tool-guided generation |
| tests/accuracy/test_single_model.sh | New script for testing individual models with specific configurations |
| tests/accuracy/test_case_ids_to_generate.json | Configuration file defining test case IDs for generation |
| tests/accuracy/install_gorilla.sh | New installation script for gorilla benchmark with updated commit hash |
| tests/accuracy/export_all_models.sh | Removed model testing logic and added new model export commands |
| extras/chat_template_examples/chat_template_devstral.jinja | New Devstral chat template with comprehensive system prompts and tool call formatting |
| demos/continuous_batching/accuracy/gorilla.patch | Updated patch for newer gorilla version with modified configuration handling |
| demos/continuous_batching/accuracy/README.md | Updated documentation with new gorilla version, installation steps, and test categories |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
tests/accuracy/test_single_model.sh
Outdated
Comment on lines
7
to
12
| --rest_port 8000 --model_repository_path /models --source_model ${model_name}-${precision} \ | ||
| --tool_parser ${tool_parser} --model_name ovms-model \ | ||
| --cache_size 0 --task text_generation | ||
|
|
||
| echo wait for model server to be ready | ||
| while [ "$(curl -s http://localhost:8000/v3/models | jq -r '.data[0].id')" != "${model_name}-${precision}" ] ; do echo waiting for LLM model; sleep 1; done |
There was a problem hiding this comment.
Continuing from the previous issue, this line also uses undefined lowercase variables ${model_name} and ${precision} instead of ${MODEL} and ${PRECISION}.
Suggested change
| --rest_port 8000 --model_repository_path /models --source_model ${model_name}-${precision} \ | |
| --tool_parser ${tool_parser} --model_name ovms-model \ | |
| --cache_size 0 --task text_generation | |
| echo wait for model server to be ready | |
| while [ "$(curl -s http://localhost:8000/v3/models | jq -r '.data[0].id')" != "${model_name}-${precision}" ] ; do echo waiting for LLM model; sleep 1; done | |
| --rest_port 8000 --model_repository_path /models --source_model ${MODEL}-${PRECISION} \ | |
| --tool_parser ${tool_parser} --model_name ovms-model \ | |
| --cache_size 0 --task text_generation | |
| echo wait for model server to be ready | |
| while [ "$(curl -s http://localhost:8000/v3/models | jq -r '.data[0].id')" != "${MODEL}-${PRECISION}" ] ; do echo waiting for LLM model; sleep 1; done |
atobiszei
reviewed
Jan 13, 2026
|
|
||
|
|
||
| docker stop ovms 2>/dev/null | ||
| docker run -d --name ovms --user $(id -u):$(id -g) --rm -p 8000:8000 -v $(pwd)/models:/models openvino/model_server:latest \ |
Collaborator
There was a problem hiding this comment.
Most probably we would need to pass image label/tag so this could be the default, but otherwise user passed arguments should be used here?
Co-authored-by: Copilot <[email protected]>
Co-authored-by: Copilot <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
🛠 Summary
CVS-179106
🧪 Checklist
``