WHAT IT IS
Command R is a model Cohere released in March 2024, with retrieval-augmented generation (RAG) and tool use as its primary design goals. Cohere was founded in Toronto in 2019 by former Google researchers and has long focused on enterprise text applications. Command R packs a 128K context into a 35B-parameter model and, in RAG settings, pairs with citation and tool-calling interfaces. It ships both as open weights and as a managed API, so enterprises can deploy it in their own environments.
Why it matters
It made retrieval augmentation, citations and tool calling first-class design goals rather than after-the-fact patches, offering a ready-made shape for enterprises wiring models into their own data — and released the weights too.
Key specs
- Parameters
- 35B (Command R)
- Context window
- 128K tokens
- Open weights
- Yes
- Released
- 2024-03
Capabilities
Conversation & Instruction Following
Understand intent across turns and act on it
Question Answering & RAG
Retrieve the evidence first, then answer from it
Tool Use & Function Calling
Let the model pick an API and fill its arguments
Text Generation
Continue a passage, one word at a time
Related concepts
Retrieval-Augmented Generation
Rather than cramming knowledge into parameters, leave it outside and look it up on demand — an open-book exam instead of a closed-book one
Transformer Architecture
Replacing word-by-word relay with a room where everyone speaks at once, so long-range dependencies are one hop away
Agents & Tool Use
Let a model do more than answer: search, call APIs, run code — and decide the next step from what came back
Comparable products
Mistral Large
2024The flagship commercial model from a European open-weights lab
Llama
2023The model family that made the open-weight route mainstream
GPT-4o
2024A natively multimodal general model, with text, image and audio through one door
Jamba
2024An open-weight model that mixes a state-space model with a Transformer