Cohere

2026-04-04: Embed v2.0, Aya Expanse 8B

Effective April 4th, 2026, the following models will be retired: embed-english-v2.0 embed-english-light-v2.0 embed-multilingual-v2.0 c4ai-aya-expanse-8b c4ai-aya-vision-8b We encourage you to migrate to our latest models, which offer enhanced performance, improved accuracy, and expanded capabilities: Embedding tasks alternatives: embed-english-v3.0 embed-multilingual-v3.0 embed-v4.0 Chat tasks alternatives: command-r7b-12-2024 command-a-03-2025 command-a-reasoning-08-2025 Our team is here to ensure a seamless transition. For questions or assistance, please don’t hesitate reaching out to us at support@cohere.com . Thank you for your continued partnership. We look forward to powering your applications with our latest AI advancements.

4 Apr 2026

Cohere Status Page: All Systems Operational

Status indicator: none. All Systems Operational

Date not specified

Cohere deprecating Rerank-v2.0 model family by December 2nd, 2024

On December 2nd, 2024, we announced the release of Rerank-v3.5 along with the deprecation of the Rerank-v2.0 model family. Fine-tuned models created from these base models are not affected by this deprecation. Best Practices: Regularly check our documentation for updates on announcements regarding the status of models. Test applications with newer models well before the shutdown date of your current model. Update any production code to use an active model as soon as possible. Contact support@cohere.ai if you need any assistance with migration or have any questions.

Date not specified

Cohere Deprecating Default Classify Endpoint via Embed Models

After January 31st, 2025, usage of Classify endpoint via the default Embed models will be deprecated. However, you can still use the Classify endpoint with a fine-tuned Embed model. By leveraging fine-tuning, you can achieve even better performance and accuracy in your classification tasks. Read the documentation on Classify fine-tuning for more information.

Date not specified

Cohere deprecating Command-R-03-2024 fine-tuned models by March 8, 2025

On March 08, 2025, we will sunset all models fine-tuned with Command-R-03-2024. As part of our ongoing efforts to enhance our services, we are making the following changes to our fine-tuning capabilities: Deprecating fine-tuning with the older Command-R-03-2024 model All fine-tunes are now powered by the Command-R-08-2024 model. Models fine-tuned with Command-R-03-2024 will continue to be supported until March 08, 2025. After this date, all requests to these models will return an error.

Date not specified

Cohere Deprecating Older Command Models and Fine-Tuning Capabilities

Effective September 15, 2025, the following deprecations will roll out. Deprecated Models: command-r-03-2024 (and the alias command-r ) command-r-plus-04-2024 (and the alias command-r-plus ) command-light command summarize (Refer to the migration guide for alternatives). For command model replacements, we recommend you use command-r-08-2024 , command-r-plus-08-2024 , or command-a-03-2025 (which is the strongest-performing model across domains) instead. Retired Fine-Tuning Capabilities: Fine-tuning for the command-light , command , command-r , classify , and rerank models is being retired. This covers both the Cohere dashboard and API. Previously fine-tuned models will no longer be accessible. Deprecated Features and API Endpoints: /v1/connectors (Managed connectors for RAG) /v1/chat parameters: connectors , search_queries_only /v1/generate (Legacy generative endpoint) /v1/summarize (Legacy summarization endpoint) /v1/classify Slack App integration Coral Web UI (chat.cohere.com and coral.cohere.com) These changes reflect our commitment to innovation and performance optimization. We encourage users to assess their current implementations and migrate to recommended alternatives. Our support team is available at support@cohere.com to assist with the transition. For detailed guidance, please refer to our migration resources and technical documentation.

Date not specified

Cohere retiring Embed v2.0 and Aya Expanse 8B models by April 4, 2026

Effective April 4th, 2026, the following models will be retired: embed-english-v2.0 embed-english-light-v2.0 embed-multilingual-v2.0 c4ai-aya-expanse-8b c4ai-aya-vision-8b We encourage you to migrate to our latest models, which offer enhanced performance, improved accuracy, and expanded capabilities: Embedding tasks alternatives: embed-english-v3.0 embed-multilingual-v3.0 embed-v4.0 Chat tasks alternatives: command-r7b-12-2024 command-a-03-2025 command-a-reasoning-08-2025 Our team is here to ensure a seamless transition. For questions or assistance, please don’t hesitate reaching out to us at support@cohere.com . Thank you for your continued partnership. We look forward to powering your applications with our latest AI advancements.

4 Apr 2026

System Status: All Systems Operational

Date not specified

Cohere releases Rerank 4.0 foundational model for ranking

Cohere has launched Rerank 4.0, their newest and most performant foundational model for ranking tasks. This represents an upgrade to their ranking capabilities with improved performance. No immediate action required, but users may want to evaluate the new model for potential performance benefits.

Date not specified

Cohere deprecating Command models and retiring fine-tuning

Cohere is deprecating several Command models including command-r-03-2024, command-r-plus-04-2024, command-light, and command summarize, requiring migration to newer alternatives like command-r-08-2024 or command-a-03-2025. All fine-tuning capabilities are being retired and previously fine-tuned models will no longer be accessible. Immediate action is required to migrate applications using deprecated models and fine-tuned models.

Date not specified

Cohere launches Command A Translate model for machine translation

Cohere launched their first machine translation model, Command A Translate, supporting 23 languages with state-of-the-art performance. This is a new capability addition that expands Cohere's model offerings. No action required for existing users.

Date not specified

Cohere launches Command A Reasoning model with 111B parameters

Cohere has launched a new hybrid reasoning model with 111 billion parameters, 256K context length, and enhanced tool use capabilities across 23 languages. The model is designed for complex agentic tasks and is accessible through the existing Command API interface. This is a new feature addition that expands available model options without requiring immediate action.

Date not specified

Cohere launches Command A Vision — first commercial multimodal model

Cohere has launched Command A Vision, their first commercial model that can process both visual and text data, expanding the Command family with enterprise-grade vision capabilities. The new model uses the existing Command API interface, making it easy for current users to integrate multimodal functionality. This is a new feature release that enhances available capabilities without requiring immediate action.

Date not specified

Cohere models now available on Oracle Cloud Infrastructure (OCI) Generative AI

Three new Cohere models are now supported on OCI: Command A 03-2025 with 150% improved throughput, Embed v3.0 with multimodal capabilities for text and images, and Rerank 3.5 supporting over 100 languages. This expansion provides OCI customers with additional enterprise-ready AI model options for search, embedding, and generation tasks.

Date not specified

Cohere Embed v4 released: Multimodal embeddings and 128k context

Cohere launched Embed v4, their most performant search model featuring unified embeddings for mixed modality input (text and images), Matryoshka embeddings, and extended 128k context length. The model achieves state-of-the-art performance in text-to-text, text-to-image, and text-to-mixed modality retrieval tasks. This is an additive release that expands Cohere's embedding capabilities without requiring immediate action from existing users.

Date not specified

Cohere releases Command A model with 111B parameters

Cohere launched Command A, a new enterprise-focused LLM with 111B parameters, 256K context length, and significantly improved inference efficiency. The model excels at tool use, RAG, agents, and multilingual tasks while requiring only two GPUs to run. Available immediately on Cohere Platform, HuggingFace, and via SDK.

Date not specified

Cohere Labs releases Aya Vision multimodal LLM

Cohere Labs announced Aya Vision, a new state-of-the-art multimodal LLM that supports multiple languages and handles both text and image inputs. The model excels at image captioning, visual question answering, text generation, and translations from text/images. This is an informational release of new capabilities with no required action for existing users.

Date not specified

Cohere releases Command R7B Arabic model with 8B parameters

Cohere has launched Command R7B Arabic, an open-weights 8-billion parameter model optimized for Arabic (MSA dialect) and English with 128,000 token context length. The model excels at enterprise tasks including instruction following, RAG, and length control while demonstrating strong Arabic language and cultural understanding. This is a new capability release requiring no immediate action from existing users.

Date not specified