GPAI Ledger › C4AI Command A Plus (Cohere) › Capture 17 Aug 2026

C4AI Command A Plus — capture 20260817T080547Z

Verify: sha256sum 83e7404c5f92a89bb44cc41bf58eb9eceecce25ebaae0accff2a9660d3094fa9.html.txt must equal the hash above (the filename IS the expected hash); ots verify 83e7404c5f92a89bb44cc41bf58eb9eceecce25ebaae0accff2a9660d3094fa9.html.ots -f 83e7404c5f92a89bb44cc41bf58eb9eceecce25ebaae0accff2a9660d3094fa9.html.txt (opentimestamps.org) proves the capture time (fresh proofs report 'pending' until bitcoin-anchored, typically within a day).

Extracted text

Machine-extracted text (layout may be lost; the authoritative content is the stored file above).

Cohere's Command A+ Model | Cohere
For AI agents: a documentation index is available at the root level at /llms.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
docs
v2 API
v2 API
Search
/
Ask AI
DASHBOARD
PLAYGROUND
DOCS
COMMUNITY
LOG IN
Guides and concepts
API Reference
Release Notes
LLMU
Cookbooks
Search
/
Ask AI
v2 API
v2 API
Guides and concepts
API Reference
Release Notes
LLMU
Cookbooks
Get Started
Welcome to Cohere
Cohere Platform
Overview
Installation
Creating a client
Quickstart
Playground
FAQs
Model Vault
Overview
Quickstart
Deploy & manage
Operate & observe
Standard Vault
Encrypted Vault
Model Vault with North
Models
An Overview of Cohere's Models
Audio
Aya
Command
Command A+
Command A
Command A Reasoning
Command A Translate
Command A Vision
Command R7B
Command R+
Command R
Embed
North
Rerank
Text Generation
Introduction to Text Generation at Cohere
Using the Chat API
Reasoning
Image Inputs
Streaming Responses
Structured Outputs
Predictable Outputs
Advanced Generation Parameters
Retrieval Augmented Generation (RAG)
Tool Use
Tokens and Tokenizers
Summarizing Text
Safety Modes
Embeddings (Vectors, Search, Retrieval)
Introduction to Embeddings at Cohere
Semantic Search with Embeddings
Multimodal Embeddings
Batch Embedding Jobs
Reranking
Going to Production
API Keys and Rate Limits
Going Live
Deprecations
How Does Cohere's Pricing Work?
Integrations
Integrating Embedding Models with Other Tools
Cohere and LangChain
LlamaIndex and Cohere
Deployment Options
Overview
SDK Compatibility
Private Deployment
Cloud AI Services
Tutorials
Cookbooks
LLM University
Build Things with Cohere!
Agentic RAG
Cohere on Azure
Responsible Use
Cohere Trust Center
Security
Usage Policy
Command A Technical Report
Command R and Command R+ Model Card
Cohere Web Crawlers
Cohere Labs
Cohere Labs Acceptable Use Policy
More Resources
Cohere Toolkit
Datasets
Improve Cohere Docs
Welcome to Cohere
Overview
Installation
Creating a client
RAG
Reranking
Semantic Search
Text Generation
Tool Use & Agents
Transcribing Audio
Playground
FAQs
Overview
Quickstart
Home Page
Creating a Vault
Managing Vaults
Monitoring
Overview
Supported Models
Calling the API
Pricing
Overview
Supported Models
Calling the API
Confidential Computing Primer
Security Model
Remote Attestation
Verifying Your Deployment
Encryption & Key Management
Compliance
Pricing
FAQ
Model Vault with North
An Overview of Cohere's Models
Cohere Transcribe
Cohere Transcribe Arabic
Aya
Aya Vision
Aya Expanse
Tiny Aya
Command A+
Command A
Command A Reasoning
Command A Translate
Command A Vision
Command R7B
Command R+
Command R
Embed
North Mini Code
Rerank
Introduction to Text Generation at Cohere
Using the Chat API
Reasoning
Image Inputs
Streaming Responses
Structured Outputs
Parameter Types in Structured Outputs (JSON)
Predictable Outputs
Advanced Generation Parameters
Basic usage
End-to-end example
Streaming
Citations
Tool Use
Basic usage
Usage patterns
Parameter types
Streaming
Citations
Tokens and Tokenizers
Summarizing Text
Safety Modes
Introduction to Embeddings at Cohere
Semantic Search with Embeddings
Multimodal Embeddings
Batch Embedding Jobs
Rerank Overview
Rerank Best Practices
API Keys and Rate Limits
Going Live
Deprecations
How Does Cohere's Pricing Work?
Integrating Embedding Models with Other Tools
Elasticsearch and Cohere
MongoDB and Cohere
Redis and Cohere
Haystack and Cohere
Pinecone and Cohere
Weaviate and Cohere
Open Search and Cohere
Vespa and Cohere
Qdrant and Cohere
Milvus and Cohere
Zilliz and Cohere
Chroma and Cohere
Cohere and LangChain
Chat on LangChain
Embed on LangChain
Rerank on LangChain
Tools on LangChain
LlamaIndex and Cohere
Overview
SDK Compatibility
Overview
Setting Up
Model Deployment
Model Deployment - AWS
Usage
Cohere on AWS
Amazon Bedrock
Amazon SageMaker
Deploy Your Own Finetuned Command-R-0824 Model from AWS Marketplace
Cohere on Azure
Cohere on Oracle Cloud Infrastructure (OCI)
Cookbooks
LLM University
Build Things with Cohere!
Cohere Text Generation Tutorial
Building a Chatbot with Cohere
Semantic Search with Cohere
Reranking with Cohere
RAG with Cohere
Building an Agent with Cohere
Agentic RAG
Routing Queries to Data Sources
Generating Parallel Queries
Performing Tasks Sequentially
Generating Multi-Faceted Queries
Querying Structured Data (Tables)
Querying Structured Data (SQL)
Cohere on Azure
Text Generation
Semantic Search
Reranking
Retrieval Augmented Generation (RAG)
Tool Use & Agents
Usage Policy
Command R and Command R+ Model Card
Cohere Web Crawlers
Cohere Labs Acceptable Use Policy
Cohere Toolkit
Datasets
Improve Cohere Docs
DASHBOARD
PLAYGROUND
DOCS
COMMUNITY
LOG IN
Light
On this page
Description
What Can Command A+ Be Used For?
General
Model Properties
Distribution and Technical Integration
Model Data and Training
Scroll to top
Models
Command
Cohere’s Command A+ Model
Copy page
Capabilities
Reasoning
Multilingual
Image Inputs
Safety Modes
Citations
Tool Use
Structured Outputs
Pricing
For both trial keys and production keys,
Command A+
 is free until rate limits are reached. Learn more about rate limits for different models and key types

here
.
Command A+ can be used in production through Cohere's
Model Vault
.
Specifications
Context Window:

128,000
 tokens
Max Output Tokens:

64,000
 tokens
Knowledge Cutoff:

April 1, 2025
API Endpoints
Model ID
command-a-plus-05-2026
Chat V2
Chat Completions
Chat V1
Try in Playground

Description

Command A+ is Cohere’s first Mixture of Experts model, bringing together the power of a strong agentic model, supporting vision and text on input, and expanding the language support to 48 languages, including all of the official European Union languages. With the MoE architecture, the model brings balance with providing a highly accurate user experience, while balancing out the need for enterprises to have a high throughput, low latency solution that minimizes the amount of GPUs required per instance, offering 1xB200 / 2xH100 support.

What Can Command A+ Be Used For?

Command A+ is excellent for:

Complex, Multimodal Agentic Tasks
: With mixed modality (vision / text) inputs, the model can autonomously take actions and interact with its environment to solve complex tasks.

Multilingual Tasks
: With support for 48 languages, combined with the model’s reasoning, agentic problem solving & vision processing capabilities, Command A+ expands not only the number of languages supported, but what enterprises across the globe can do with Cohere’s models.

Balancing Accuracy & Efficiency
: Providing support for deploying the model with as few as a single B200 or two H100 chips, and with vast improvements in reasoning, image processing, and agentic tasks over the rest of the Command A family, Command A+ is the fastest and most performant model in the Command A family by far.

There’s more to be said about token budgets, enabling and disabling the
thinking
 operation, etc., which can be found in our dedicated
Reasoning guide
.

General

Name of the model provider
: Cohere Inc.

Release Date
: May 20, 2026

Model dependencies
: N/A

Contact
:
support@cohere.com

Model Properties

Model architecture
: Command A+ is a sparse mixture-of-experts model.

Input modalities
: Text, image

Output modalities
: Text

Model Size
: 218B total, 25B active

Distribution and Technical Integration

Cohere Deployment
:
Deployment options
 are subject to terms of use applicable to the deployment type or platform. Usage is subject to limitations set out in commercial agreements with customers, which includes Cohere’s
Usage Policy
.

Open Source Deployment
: Command A+ is available under an Apache 2.0 License on
Hugging Face
.

Technical Integration and Required Software (if any)
: For Cohere Deployments, see our
Installation Guide
 on getting started with Command A+. For Open Source Deployments, instructions are available on the
HuggingFace page
.

Required Hardware
: 1× B200 at W4A4 or 2× H100s at W4A4

Model Data and Training

Pre-training
: The model was trained on a large-scale collection of unlabelled data using self-supervised learning to learn general patterns, syntax, and semantics across multiple languages.

Post-training
: During post training, labelled datasets were used with supervised fine-tuning (SFT) and reinforcement learning (RL) techniques to maximize the model’s performance over a wide spectrum of domains and capabilities.

Data
: Across both pre-training and post-training, various data sources and modalities are used, which include text and image content sourced from publicly available information, proprietary datasets developed or generated by Cohere with human annotation or through automated means, including synthetic data, and datasets sourced from specialized data vendors. In most cases, Cohere customers use Cohere models in their own environments, meaning Cohere has no access to inputs submitted to its models and does not use them for model training. De-identified data from the use of Cohere models on Cohere-hosted environments (e.g. user inputs) may be used in limited circumstances where permitted by user controls and Cohere’s relevant terms of service.

Training Data Processing
: Cohere employs various techniques to ensure data quality and suitability for training, including deduplication, filtering for toxic or harmful content, and quality filtering.

EU AI Act Article 53(1)(d)
:
Public Summary of Training Content

Was this page helpful?
Yes
No
Edit this page
Previous
Command A
Our largest, most performant model, ideal for building enterprise agents with a low compute footprint.
Next
Built with