Multilingual multimodal embedding model optimized for low latency and high throughput in interactive search and high-volume retrieval. Shares an embedding space with Embed 5 Pro.
Multilingual multimodal embedding model optimized for enterprise document retrieval, including financial documents and parsed PDFs. Shares an embedding space with Embed 5 Fast.
Parse v5.0 is Cohere's document parsing model for extracting ordered text, tables, images, captions, and layout-aware blocks from document images.
Cohere Transcribe Arabic is a multidialectal Arabic-English automatic speech recognition model optimized for far-field audio and code-switching.
North Mini Code is Cohere's first agentic coding model and the debut of its North family. A sparse mixture-of-experts model with 30B total parameters and 3B active, it is optimized...
Command A+ is Cohere's updated Command-family multimodal model with image understanding, a 128k context window, and stronger instruction-following and agentic performance.
Cohere Transcribe is a multilingual 2B-parameter automatic speech recognition model for audio-to-text transcription.
Tiny Aya Earth is a language model from Cohere for chat, writing, analysis, and general-purpose automation. It is a smaller-footprint option suited to tighter latency or deployment budgets.
Tiny Aya Fire is a language model from Cohere for chat, writing, analysis, and general-purpose automation. It is a smaller-footprint option suited to tighter latency or deployment budgets.
Tiny Aya Global is a language model from Cohere for chat, writing, analysis, and general-purpose automation. It is a smaller-footprint option suited to tighter latency or deployment budgets.
Tiny Aya Water is a language model from Cohere for chat, writing, analysis, and general-purpose automation. It is a smaller-footprint option suited to tighter latency or deployment budgets.
Cohere's AI search foundation model for enhancing the relevance of information surfaced within search and RAG systems. Use it to rescore retrieved passages, products, or documents before final ranking or generation.
Cohere's AI search foundation model for enhancing the relevance of information surfaced within search and RAG systems. Use it to rescore retrieved passages, products, or documents before final ranking or generation.
Command A Translate is a translation model from Cohere for converting text between languages. It is aimed at multilingual translation and localization workflows.
Command A Reasoning is a reasoning model from Cohere for multi-step analysis, math, and tool-driven tasks.
Command A Vision is a multimodal model from Cohere that can work across text and visual inputs.
Embed V4 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at semantic search, retrieval, clustering, and RAG pipelines.
Command A is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.
C4AI Aya Vision 32B is a multimodal model from Cohere that can work across text and visual inputs.
Command R7B (12-2024) is a small, fast update of the Command R+ model, delivered in December 2024. It excels at RAG, tool use, agents, and similar tasks requiring complex reasoning...
Command R 7b is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
C4AI Aya Expanse 32B is a language model from Cohere for chat, writing, analysis, and general-purpose automation.
Rerank V3.5 is designed to reorder search results for improved relevance. It supports multi-aspect and semi-structured data reranking over 100+ languages.
command-r-08-2024 is an update of the [Command R](/models/cohere/command-r) with improved performance for multilingual retrieval-augmented generation (RAG) and tool use. More broadly, it is better at math, code and reasoning and...
command-r-plus-08-2024 is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint...
Command R is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Command R (2024-08-30) is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Command R+ (2024-08-30) is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Rerank English V3 is designed to reorder English-language search results for improved relevance, including semi-structured data.
North-Mini-Code-1.0 is Cohere's first North-family code agent model, a 30B total / 3B active Mixture-of-Experts model built for agentic coding, terminal workflows, and local deployment.
Command R Plus is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Embed English Light V3 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at English-language retrieval, clustering, and semantic search.
Embed English V3 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at English-language retrieval, clustering, and semantic search.
Embed Multilingual Light V3 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at multilingual retrieval and semantic similarity tasks.
Embed Multilingual V3 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at multilingual retrieval and semantic similarity tasks.
Rerank Multilingual V3 is designed to reorder search results for improved relevance. It supports multi-aspect and semi-structured data reranking over 100+ languages.
Command R+ (2024-04-04) is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Command R (2024-03-11) is an update of the [Command R+](/models/cohere/command-r-plus) with roughly 50% higher throughput and 25% lower latencies as compared to the previous Command R+ version, while keeping the hardware footprint.
Command is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.
Command Light is an open-weights 111B parameter model with a 256k context window focused on delivering great performance across agentic, multilingual, and coding use cases.
C4AI Aya Vision 8B is a multimodal model from Cohere that can work across text and visual inputs.
C4AI Aya Expanse 8B is a language model from Cohere for chat, writing, analysis, and general-purpose automation.
Embed English Light V2.0 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at English-language retrieval, clustering, and semantic search.
Embed English V2.0 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at English-language retrieval, clustering, and semantic search.
Embed Multilingual V2.0 is an embedding model from Cohere for vector search and semantic retrieval workflows. It is aimed at multilingual retrieval and semantic similarity tasks.