Cohere LLMs
Enterprise-focused LLM provider known for RAG-oriented models and embeddings.
Cohere positions Command R+ as a RAG-first model with native citations and structured tool use. Its embedding models (Embed v3) are widely deployed for retrieval pipelines. Pricing skews enterprise — quotes via sales for high volume — but list prices below apply to direct API use.
Founded: 2019 · HQ: Toronto, Canada · Docs: docs.cohere.com ↗
All Cohere models
| Model | Family | Context | Input $/M | Output $/M | Released | Status |
|---|---|---|---|---|---|---|
| Command R+ | Command R | 128K | $2.50 | $10.00 | 2024-04-04 | active |
Comparisons with other providers
The most-searched comparisons involving Cohere models:
Working with the Cohere API
Documentation lives at https://docs.cohere.com. Before paying for any call, count input tokens with the appropriate counter:
- OpenAI Token Counter — for GPT-* and o-series models
- Claude Token Counter — for Anthropic Claude models
- Gemini Token Counter — for Google Gemini models