Command R+: Pricing, Context, and Benchmarks
Command R+ is Cohere's Command R model released on 2024-04-04. It runs on a 128K-token context window, priced at $2.50/M input and $10.00/M output. RAG-first model with native citations and tool calling. Strong choice for enterprise retrieval.
Specifications
| Provider | Cohere |
|---|---|
| Family | Command R |
| Released | 2024-04-04 |
| Status | active |
| Context window | 128K tokens |
| Max output | 4K tokens |
| Modalities | text |
| Capabilities | function calling json mode streaming rag citations |
| Tokenizer | cohere |
| API endpoint | https://api.cohere.com/v2/chat |
| Documentation | Official docs ↗ |
| Last updated | 2026-05-29 |
What this model is good for
Based on benchmark scores and capability fit, Command R+ ranks well for:
- Document Q&A / RAG — rank #1. Native citations for grounded answers. Built for RAG.
Cost calculator
For 1,000 tokens of input and 1,000 tokens of output per call, this model costs $0.0125 per call. For typical usage:
| Workload | Tokens / call (in/out) | $/call | $/1K calls | $/1M calls |
|---|---|---|---|---|
| Short Q&A | 200 / 100 | $0.001500 | $1.50 | $1,500 |
| Standard chat | 1K / 500 | $0.007500 | $7.50 | $7,500 |
| RAG with retrieval | 4K / 500 | $0.015000 | $15.00 | $15,000 |
| Long doc summary | 20K / 1K | $0.060000 | $60.00 | $60,000 |
| Long context (100K input) | 100K / 2K | $0.265000 | $265.00 | $265,000 |
For more precise estimates including caching discounts, use the LLM API Cost Calculator.
Common comparisons
- Command R+ vs GPT-4o Side-by-side specs, pricing, and benchmarks
- Command R+ vs GPT-4.1 Side-by-side specs, pricing, and benchmarks
- Command R+ vs o3 Side-by-side specs, pricing, and benchmarks
- Command R+ vs Claude Sonnet 4.6 Side-by-side specs, pricing, and benchmarks
API setup
curl https://api.cohere.com/v2/chat \
-H "Authorization: Bearer $API_KEY" \
-H "Content-Type: application/json" \
-d '{"model": "command-r-plus", "messages": [{"role": "user", "content": "Hello"}]}'See the official documentation for the full request/response schema. For pre-flight token counting before paying, use the appropriate token counter.