Gorilla
First page
Paper summary
A fine-tuned LLaMA-based model that surpasses GPT-4 on API call generation.
Ask this paper
01
API-specialized LLM: Specifically trained on massive API documentation corpora to produce correct API calls for TensorFlow Hub, HuggingFace, and PyTorch Hub.
02
Beats GPT-4 on APIs: Outperforms GPT-4 on writing correct API calls - a narrow but important capability for tool use.
03
Hallucination reduction: Major reduction in hallucinated API names and parameters compared to general-purpose LLMs.
04
Tool-use LLM research: Established that specialized LLMs can meaningfully beat generalists at narrow capabilities - informing the later ecosystem of task-specialized models.