LLM Reference

Vicuna 13B V1.5

About

Vicuna 13B V1.5, developed by LMSYS, is a large language model designed for chat assistance, fine-tuned from Llama 2 using the transformer architecture. It underwent supervised instruction fine-tuning with around 125,000 conversations from ShareGPT. The model excels in text generation, conversational dialogue, and Q&A tasks, primarily intended for research. While its responses are high-quality and polite, it may lack domain-specific insights, exhibit limited contextual understanding, and display inherited biases. Quantized versions of the model provide trade-offs between size, speed, and accuracy 129.

Capabilities

MultimodalFunction CallingTool UseJSON Mode

Providers(1)

ProviderInput (per 1M)Output (per 1M)Type
Together AI API$0.3$0.3
Serverless

Specifications

FamilyVicuna
Parameters13B
Context2K
ArchitectureDecoder Only
Specializationgeneral