Activefrontierllm Proprietary

Command R

by Cohere· Released March 2024· Cutoff August 2023

Command R is Cohere's flagship large language model optimized for enterprise use cases, including retrieval-augmented generation (RAG) and tool use. It balances high performance with efficiency, offering strong multilingual support and a 128K context window. Command R is designed to be a versatile workhorse for production applications.

Official Site API Docs

Input cost

$0.50 per 1M tokens

Output cost

$1.50 per 1M tokens

Context window

128K tokens

Max output

4096 tokens

Modalities

text

Parameters

35B

License

proprietary

Capabilities

Function CallingTool UseRetrieval-Augmented Generation (RAG)Multilingual SupportStreamingJSON ModeCode GenerationLong Context

Best For

Enterprise applications requiring reliable RAG, tool use, and multilingual capabilities with a large context window.

Strengths

Excellent at retrieval-augmented generation (RAG) tasks
Strong multilingual performance across 10+ languages
Efficient and cost-effective for production use
Reliable function calling and tool use

Limitations

Not as strong in creative writing or open-ended generation as some larger models
Limited vision capabilities (text-only)
Smaller parameter count may limit performance on complex reasoning compared to frontier models like Command R+
No image generation or multimodal support

Use Cases

Enterprise knowledge base Q&A with RAG

Multilingual customer support chatbots

Automated document summarization and analysis

Code generation and assistance

Data extraction and structured output generation

Tool-using agents for workflow automation

Content moderation and classification

Improvements Over Previous Model

Introduced as a new model family, not an incremental update
Optimized for RAG and tool use from the ground up
128K context window, significantly larger than previous Command models (which had 4K-8K)
Multilingual support expanded to 10+ languages
Lower pricing compared to previous Command models
Improved instruction following and reliability for enterprise use

Back to all models

Improvements Over Previous Model

Introduced as a new model family, not an incremental update

Optimized for RAG and tool use from the ground up

128K context window, significantly larger than previous Command models (which had 4K-8K)

Multilingual support expanded to 10+ languages

Lower pricing compared to previous Command models

Improved instruction following and reliability for enterprise use