LiteLLMModels

Cohere model pricing

As of October 10, 2026, 20 Cohere models are available through LiteLLM. The most recently added are Embed V5.0 Fast, Embed V5.0 Pro, C4ai Aya Expanse 32B. Prices below are Cohere's list prices.

How to call Cohere with LiteLLM

Cohere models

ModelLabPriceContextProvidersAdded
Embed V5.0 Fast–$0.08 per 1M tokens128K1Oct 2026
Embed V5.0 Pro–$0.12 per 1M tokens128K1Oct 2026
C4ai Aya Expanse 32BCohere$0.50 / $1.50 per 1M tokens128K1Sep 2026
Command A PlusCohereFree128K2Sep 2026
Rerank V4.0 FastCohere$0 per 1M input tokens32K1Sep 2026
Rerank V4.0 ProCohere$0 per 1M input tokens32K1Sep 2026
Parse V5.0–––1Sep 2026
Embed Multilingual Light V3.0Cohere$100 per 1M tokens5122Nov 2025
Embed V4.0Cohere$0.12 per 1M tokens128K4May 2025
Command ACohere$2.50 / $10 per 1M tokens256K3Apr 2025
Command R7BCohere$0.0375 / $0.15 per 1M tokens128K1Jan 2025
Rerank V3.5Cohere$0 per 1M input tokens4K2Dec 2024
Rerank English V3.0Cohere$0 per 1M input tokens4K1Oct 2024
Rerank Multilingual V3.0Cohere$0 per 1M input tokens4K1Oct 2024
Embed English Light V3.0Cohere$0.10 per 1M tokens5122Aug 2024
Embed English V3.0Cohere$0.10 per 1M tokens5122Aug 2024
Embed Multilingual V3.0Cohere$0.10 per 1M tokens5122Aug 2024
Command R PlusCohere$2.50 / $10 per 1M tokens128K4Apr 2024
Command RCohere$0.15 / $0.60 per 1M tokens128K3Mar 2024
Command NightlyCohere$1 / $2 per 1M tokens4K1–