ModelPriceWatch.com
Last scan 2026-09-22 Models tracked 267 Providers 35 Cheapest paid Granite 4.0 H Micro $0.017/Mtok in Every price links to its source

Command A

by Cohere · 111B parameters

Current flagship mid tier

Today's price · per 1M tokens

Input

$2.50

per 1M tokens

Output

$10.00

per 1M tokens

Blended

$4.38

blended $/1M — 3:1 weighted input:output

How this price is scoped: Cohere's list price for Command A, read 2026-08-22 on the first-party model card at docs.cohere.com/docs/command-a: Input $2.5 / 1M tokens, Output $10 / 1M tokens, printed beside the 256,000-token context window and the model ID command-a-03-2025 that this row carries. We cite that page rather than cohere.com/pricing because Cohere's public rate card no longer prices Command A at all: with all four of its tabs opened, the "Generative models" tab prices only Command A+ ($0 / Free under Apache 2.0 — a different, open-weights SKU), Command R ($0.15 / $0.60) and Command R7B ($0.0375 / $0.15). The rate card's legacy "For existing customers" block does print $2.50/$10, but for Command R+ 08-2024 — the same two figures as this row, which is a coincidence and not a source for it. Command A itself is current, not withdrawn: docs.cohere.com/docs/models lists command-a-03-2025 with status Live.

Source: official Cohere pricing · read MODELPRICEWATCH.COM · 2026-09-22

Price receipt

We read Cohere's own pricing page on and found Command A listed with a price on the same row — that read is where the number above comes from. We keep the page text we read; its content hash is c6b69e5975.

Overview

Cohere's flagship model. $2.50/$10 per 1M tokens, 256k context window. Priced on Cohere's Command A model card — the cohere.com/pricing rate card no longer lists this SKU. See the price note.

Capabilities

struck through = not supported
Input 1/4
Text Image Audio Video
Output
Text
Features 1/9
Prompt caching Reasoning Coding Fast inference Long context Open weights Multimodal Web search Realtime

Benchmark performance

accuracy % · higher is better
Percentile vs all tracked models 12.5th1 independent measurement
Percentile per $/Mtok 2.9
GPQA Diamond vendor
50%
SWE-Bench Verified vendor
25%
Humanity's Last Exam vendor
15%
ARC-AGI 2 vendor
12%
AIME 2025 vendor
45%
MMMLU vendor
74%
BFCL vendor
60%
HumanEval vendor
78%
MATH 500 vendor
62%

Every per-benchmark score we hold for this model is reported by its own vendor, so none is counted toward an accuracy average or any ranking. The percentile above is its standing across the independent composites below.

Independent composite scores
  • LMSYS Chatbot Arena: 1353.6 ELO (Human preference)
How it stacks up
  • Ranks #155 of 181 comparably-measured models by percentile score, across 1 independent measurement
  • Ranks #171 of 181 comparably-tested models by normalized performance per dollar

150 tokens/sec output

Source: Cohere model card, LMSYS Chatbot Arena (UC Berkeley) · updated · See full rankings →

Specifications

Provider
Cohere
Context window
256K tokens
Modality
text
Parameters
111B
Open source
No — proprietary
Released
Status
Current
Last updated
Tags
flagship

Availability verified: listed on Cohere's own page