Checked against the OpenRouter public models feed. Prices are copied, not estimated.

The Best Model ForWhich AI model to use, from ten labs.

Cohere

Short answer

Cohere's feed is smaller: Command A+ for enterprise tool use, the older open-weight Command A, and the small Command R7B for retrieval and tools. Command A is listed at a higher token price than Command A+, so open weights are not a discount here. Embed and rerank models are listed too.

Lab site · Filter the full catalogue to Cohere

What the top models are good for

1. Command A+

Flagship · $0.30 in · $1.50 out · 192,000 tokens

Guide note · 19 Nov 2026

OpenRouter describes Command A+ as Cohere's flagship for enterprise agent workflows: text and image input, a 192,000-token window, tool calling, and a public weights id. The window fits retrieved pages. It does not fit an entire unfiltered archive.

  • Tools and structured calls
  • Image input
  • Open weights id on the flagship row

What it is good for

Turn a photo of a form into a tool call

Image input plus strict tool schemas, which the listing calls out, is a form-to-fields job. Reject calls that fail the schema.

Command A+ is listed at $0.30 input and $1.50 output per million tokens, with a 192,000-token context window.

Answer from a policy handbook you already retrieved

Put the retrieved pages in the prompt. 192,000 tokens is a retrieval window, not a reason to skip retrieval. Ask the answer to cite the page you supplied.

Command A+ is listed at $0.30 input and $1.50 output per million tokens, with a 192,000-token context window.

The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.

2. Command A

Open weights · $2.50 in · $10.00 out · 256,000 tokens

Guide note · 19 Nov 2026

Command A is the older open-weight 111B model. The listing gives it a 256,000-token window and points at agentic, multilingual, and coding use. On this feed its token price is higher than Command A+. Open weights are not a discount here. This row does not list image input. The published cutoff is 2024-08-31.

  • Open weights you can host
  • 256k context
  • Multilingual and coding, per the listing

What it is good for

Host a general model with a 256k window

Pick Command A when the requirement is the weights, not the lowest API bill. The cutoff is 2024, so attach anything newer.

Command A is listed at $2.50 input and $10.00 output per million tokens, with a 256,000-token context window.

A multilingual draft a person will review

Multilingual use is in the listing. The reviewer is not optional for anything a customer will see.

Command A is listed at $2.50 input and $10.00 output per million tokens, with a 256,000-token context window.

The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.

3. Command R7B (12-2024)

Budget · $0.0375 in · $0.15 out · 128,000 tokens

Guide note · 19 Nov 2026

Command R7B (December 2024) is the small, cheap Cohere model. The listing says it is for RAG, tool use, and agents. It is older, text only, and open weights. Test it on your data before you trust a queue to it.

  • Low rate
  • RAG and tools, per the listing
  • Open weights

What it is good for

Answer from three retrieved chunks

Short context you chose yourself is the RAG job the listing names. Do not dump the whole handbook in and hope.

Command R7B (12-2024) is listed at $0.0375 input and $0.15 output per million tokens, with a 128,000-token context window.

Bulk answers over short policies

A small model and a short policy is a cost decision. Keep a person on the ones the model marks as unsure.

Command R7B (12-2024) is listed at $0.0375 input and $0.15 output per million tokens, with a 128,000-token context window.

The tasks are editorial illustrations, not measured benchmarks. The prices in them are read from today's feed.

Every current row

12 Cohere rows on the feed.
ModelKindContextInputOutputListedGuide
Command A Chat 256,000 $2.50 $10.00 13 Mar 2025 Reviewed
Command A+ Chat 192,000 $0.30 $1.50 22 Sep 2026 Reviewed
Command R (08-2024) Chat 128,000 $0.15 $0.60 30 Aug 2024 New, not yet reviewed
Command R+ (08-2024) Chat 128,000 $2.50 $10.00 30 Aug 2024 New, not yet reviewed
Command R7B (12-2024) Chat 128,000 $0.0375 $0.15 14 Dec 2024 Reviewed
Embed 5 Fast Embeddings 128,000 $0.08 $0 3 Nov 2026 New, not yet reviewed
Embed 5 Pro Embeddings 128,000 $0.12 $0 3 Nov 2026 New, not yet reviewed
Rerank 4 Fast Rerank 32,768 $0 $0 6 Apr 2026 New, not yet reviewed
Rerank 4 Pro Rerank 32,768 $0 $0 6 Apr 2026 New, not yet reviewed
Rerank 5 Fast Rerank 32,768 $0 $0 3 Nov 2026 New, not yet reviewed
Rerank 5 Pro Rerank 32,768 $0 $0 3 Nov 2026 New, not yet reviewed
Rerank v3.5 Rerank 4,096 $0 $0 5 Apr 2026 New, not yet reviewed

Questions

Which Cohere model should I start with?

For most people, start with Command A+, then drop to the next model on this page when the task is routine. Read the cheaper-model note under each example.

Are the examples benchmarks?

No. They are concrete tasks used to explain fit. The prices next to them are live figures from the feed. The judgment is a guide note.