LLM Reference

Mistral 7B OpenOrca

About

The OpenOrca Mistral 7B is a high-performance open-source AI model that builds upon the Mistral 7B base, known for its outstanding efficiency relative to its compact size 1. Utilizing a transformer architecture, it incorporates innovations like sliding window attention and a rolling buffer cache to reduce memory use and enhance processing on conventional GPUs 8. The model excels in tasks such as text and code generation, question answering, and dialogue, often outperforming larger counterparts on various benchmarks 2. Trained on a refined subset of the OpenOrca dataset, it maintains a balance between quality and performance, despite some limitations like data bias and limited domain knowledge 11.

Capabilities

MultimodalFunction CallingTool UseJSON Mode

Providers(2)

ProviderInput (per 1M)Output (per 1M)Type
Together AI API$0.2$0.2
Serverless
Fireworks AI Platform
Provisioned

Specifications

FamilyOpenOrca
Released2023-10-03
Parameters7B
Context8K
ArchitectureDecoder Only
Specializationgeneral