Command A+

Cohere's first mixture-of-experts model, released May 2026 with 218B total parameters (25B active), combining vision, agentic work, optional reasoning and translation across 48 languages. Apache 2.0.

Params: 218B, 25B active
Context: 128K tokens
Max output: 64K
Released: May 2026
License: Apache 2.0

Visit Cohere Compare vs other LLMs →

Hosted API pricing

How we price

Order

Providers are listed cheapest input rate first, then by output rate, a tie going to the creator's own listing. A provider that lists the model without a published rate sits at the end.

Prices

Base-tier, on-demand rates in USD per 1M tokens, converted at ECB reference rates where a provider publishes in another currency. Cached-input and batch rates show only where a provider publishes them. "Cheapest" marks the one provider with the lowest input and output pair; a tie wears no badge.

Cost

Input rate times the input tokens plus output rate times the output tokens, at the monthly volume set above the table. Cached-input, batch, long-context and reasoning-token billing are not modelled.

Transparency and funding

Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.

Every provider serving Command A+, cheapest input first. Set your monthly volume to see what each would bill.

Input M Output M

Every provider serving Command A+, with its price per 1M input and output tokens and what the volume above would cost
Provider	Input /1M	Output /1M	Cost at 10M in + 2M out	Link
Cohere Creator	Open weights, no hosted price			Visit website

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

Command A+ needs about 123 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 8× RTX 3090 at about $1,440 a month.

Single node ≤ 8

What it costs to self-host Command A+ at each precision: the memory it needs, the cheapest rental that holds it, that rental's month and the API volume it pays off at
Precision	Memory	Cheapest fit (32K context)(Full 128K context)	Cost /mo
4-bit INT4 / FP4	123 GB126 GB	8× RTX 3090	$1,440$1,440
8-bit FP8 / INT8	221 GB224 GB	8× RTX A6000	$3,398$3,398
16-bit FP16 / BF16	439 GB442 GB	8× A100	$10,714$10,714

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. No guarantee of runtime support, usable performance, or that a matching quantized build exists. Pricing methodology.

Packet·ai

B200 @ $3.75/h on demand

No contract, no spot interrupts, no cluster minimums

Full root SSH, web dashboard, CI/CD ready

Reserve Now

Capabilities

What Command A+ accepts and can do, as published by Cohere.

Text Accepts and generates natural-language text.
Image Accepts images as input alongside text.
Reasoning Thinks before it answers, always on or as a switchable mode.
Function calling Connect to external tools, APIs, and systems.
Structured output Return responses in structured formats like JSON.

Common questions

What is Command A+ good for?

Complex agent workflows that mix images and text, in 48 languages, with optional reasoning. Apache 2.0 weights allow commercial self-hosting, on a multi-GPU node.

When is Command A+ not a good fit?

Audio or video input: it takes text and images. Self-hosting takes a multi-GPU node, well above the dense Command A models in the same lineup.

What is the cheapest way to run Command A+?

No provider we track hosts it; self-hosting on the cheapest rental that fits (8x RTX 3090, $2.00 per hour) costs about $1,440 a month.

Can I self-host Command A+?

Yes. The weights are Apache 2.0 licensed. At 4-bit it needs about 123 GB of GPU memory, which starts at roughly $1,440 a month on the cheapest rental that fits.