Command A+
Neural Network
Command A+ is Cohere's flagship model for agentic workflows with 192K context and native tool use.
Specifications
- Input
- text, images, documents
- Output
- text
- Context size
- 192 000 tokens
- Max answer length
- 64 000 tokens
- Reasoning effort
- low, medium, high
- Capabilities
- reasoning, function calling
Prompt cost
(per 1M tokens)
Answer cost
(per 1M tokens)
How it works Command A+?
Command A+ is Cohere's flagship model designed for enterprise agentic scenarios. It accepts text and images, supports up to 192K tokens of context, and natively calls tools using strict schemas, while also returning structured responses—the foundation for multi-step agents and internal service integrations. In BotHub, the model is accessible from Russia without a VPN or foreign card: you pay in rubles only for the tokens used, not for a subscription. Other models in the catalog are available in the same window, allowing you to compare responses or switch without rewriting integrations, thanks to a unified OpenAI-compatible API. Chats are encrypted, and companies can access contracts, invoices, electronic document management (EDO), and an admin panel with limits. It is useful when you need to build an agent that interacts with corporate tools and APIs; analyze large document packages or long threads without splitting them; extract data from scans, diagrams, and screenshots; or get strictly structured responses for databases or business processes.Frequently asked questions about Command A+
You can use generated results for commercial purposes. You own all rights to the content you create. The only restriction: make sure your prompt does not include copyrighted third-party material. You are responsible for respecting the rights to any input data.
This is a Cohere text model that accepts not only text but also documents and images. It is suitable for analyzing long materials, analytics, coding and debugging, as well as tasks requiring sequential reasoning. Context is 192,000 tokens.
Up to 64,000 tokens in a single response—enough for a lengthy article, technical documentation, detailed contract analysis, or a large code module. If the task is larger, break it down: the 192,000-token context allows you to keep the entire project in the dialogue.
Yes, the model's reasoning mode is confirmed. Before the final answer, it builds a chain of steps, which significantly helps with math, logic, data analysis, and code debugging. It is not needed for simple questions, but for complex tasks, the answer is more accurate.
Yes, the model can call functions and output structured JSON, so it can be connected to external services, databases, and agentic scenarios. In BotHub, it is available via a unified OpenAI-compatible API: you can switch to another model without rewriting the integration.
Images and documents—yes, this is confirmed: you can send a screenshot, diagram, table, or PDF and ask to analyze the content. There is no information about audio and video in our data, so for such files, it is safer to use a specialized model in BotHub.
We do not have verified data on quality by language, so we will not make any promises—it is better to evaluate it on your own task. In BotHub, this is done in a couple of clicks: the interface is in Russian, payment is via Russian card without VPN, and you only pay for the tokens actually used.