bfcl-submission

BFCL Submission Model

This model is a submission for the Berkeley Function-Calling Leaderboard (BFCL), designed to evaluate LLM function-calling capabilities.

Model Details

• Model Type: Causal Language Model
• Base Model: microsoft/DialoGPT-medium
• Mode: fc (native function-calling)
• Parameter Count: ~345M parameters
• License: Apache 2.0

Function Calling Capabilities

The model can execute the following functions:

1. web_search: Search the web for information
2. get_weather: Get current weather information
3. calculate: Perform mathematical calculations
4. store_memory: Store information in memory
5. retrieve_memory: Retrieve information from memory

Usage

The model is designed to be used with the BFCL evaluation framework. The main entry point is the process_message function in handler.py.

from handler import process_message

# Process a message
result = process_message("Hello, how are you?")
print(result)

Evaluation

This model will be automatically evaluated by the BFCL team using their pinned evaluator version.

Repository Information

• GitHub: https://github.com/sweetdream0530/bfcl_submission
• HuggingFace: https://huggingface.co/sweetdream0530/bfcl-submission
• BFCL Submission: Use this HuggingFace URL for BFCL evaluation

License

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.