Wispr Flow - Don’t type, just speak
Mini tools| Model | Input | Cached input | Output | Unit |
|---|---|---|---|---|
|
Command R
Caldera Labs
|
$0.15 | - | $0.6 | per 1M tokens |
|
Command R7B
This model
Cohere
|
$0.0375 | - | $0.15 | per 1M tokens |
Command R7B
By Cohere
Model family: Command
Command R7B is a 7B parameter large language model built for enterprise use cases requiring speed, cost efficiency, and low compute overhead. With a 128K context window, it delivers strong performance on retrieval-augmented generation (RAG), tool use, structured outputs, and multi-step agentic workflows. The model supports image inputs, multilingual text, and complex reasoning, making it suitable for diverse real-world tasks. Its compact architecture enables deployment on consumer GPUs and CPUs for on-device inference. Command R7B is particularly strong at high-throughput chatbot applications, code assistance, and agent tasks that require information-seeking and multi-step reasoning. It avoids unnecessary tool calls, improving reliability in practical applications. Accessible via the Cohere API and SDK using model ID command-r7b-12-2024, and available as open weights under a CC-BY-NC license.
Multimodal
Gen 3
Overview
Command R7B is Cohere's smallest and fastest model in the Command R family, with a 128K context window and 7B parameters. Designed for high-throughput, latency-sensitive applications, it supports image inputs, multilingual text, tool use, RAG, and multi-step agentic tasks. Its compact size enables on-device inference on consumer GPUs and CPUs.
Pricing
Compare Command R7B with other models listed in the same vendor pricing tiers and context lengths.
Standard
About Cohere
Cohere is a technology company specializing in Artificial Intelligence, Natural Language Processing, and Machine Learning solutions.
Industry:
Artificial Intelligence
Company Size:
900
Location:
Toronto, Ontario, CA
Website:
cohere.com
Last updated: June 22, 2026
