{T}
TokenMath.net
Official Provider HubToronto, Canada

Cohere API Pricing & Token Economics

Cohere builds foundation models purpose-engineered for enterprise RAG, search grounding, and multi-step tool execution (Command R+ and Command R). Cohere models feature native citation grounding, 256k vocabulary tokenizers, and 90% prompt caching discounts on document prefixes.

Prompt Caching Policy

90% read discount on cached documents and prompt prefixes ($0.25/1M on Command R+).

Batch / Asynchronous Rules

Standard API execution rules. Refer to individual endpoint rate limits for asynchronous processing.

Special Billing Conditions

Built-in document citation generator links output claims directly to retrieved text chunks.

Active Cohere Models(2 models verified)

All 30 Models →
ModelContextInput / 1MCached / 1MOutput / 1MBatch DiscountActions
Command R+ (08-2024)
command-r-plus-08-2024
128k$2.50$0.25$10.00N/ASpecs →
Command R (08-2024)
command-r-08-2024
128k$0.15$0.015$0.60N/ASpecs →

Calculate your monthly Cohere bill

Simulate prompt caching hit rates, batch discounts, and output token costs with zero server logging.

Open Cost Calculator →