gorilla-llm/gorilla-7b-hf-delta-v1
Gorilla-7b-hf-delta-v1 is a 7 billion parameter instruction-tuned language model developed by Gorilla LLM (UC Berkeley), based on the LLaMA architecture with a 4096 token context length. This model specializes in generating accurate API calls from natural language queries, specifically for Hugging Face APIs. It is designed to reduce hallucination in API invocation and can output directly executable code snippets for various tasks like translation and object detection.
Loading preview...
Gorilla-7b-hf-delta-v1: API Invocation LLM
Gorilla-7b-hf-delta-v1 is a 7 billion parameter language model, fine-tuned from LLaMA, designed to enable Large Language Models (LLMs) to effectively use tools by invoking APIs. Developed by Gorilla LLM (UC Berkeley), this model excels at translating natural language queries into semantically and syntactically correct API calls, significantly reducing hallucination in the process. It has been specifically trained to reliably use over 1,600 Hugging Face APIs.
Key Capabilities
- API Generation: Translates natural language requests into executable API code snippets.
- Reduced Hallucination: Demonstrates improved accuracy in generating valid API calls compared to general-purpose LLMs.
- Hugging Face API Focus: Optimized for invoking a wide range of Hugging Face APIs for tasks such as translation, image processing, and more.
- Direct Code Output: Generates code that can be directly integrated into workflows, as shown in examples for
transformerspipeline usage.
Good For
- Developers looking to integrate LLMs with external tools and services via APIs.
- Automating tasks that require interaction with Hugging Face models and libraries.
- Building applications where an LLM needs to dynamically generate and execute code for specific functions.
- Research into tool-use and API invocation capabilities of large language models.