Skip to main content
AI Choice Engine

Cohere

Cohere Command A+

Cohere Command A+ is a Cohere model with 128K context, $0.80 in / $3.20 out per million tokens, last verified October 2, 2026. Open weights: yes. Public-board scores are linked to the publisher; vendor-only figures stay in the watch-out, not on the board.

Command A+ is Cohere's Apache-declared open-weight 218B/25B sparse MoE for enterprise text, image and agentic workloads; its W4A4 checkpoint is distinct from a hosted service contract.

Recorded profile verification October 2, 2026Individual provider fields may change

AI model specification details
SpecificationCohere Command A+
ProviderCohere
TierBalanced
Context window128K
Max output64K
Input / 1M tokens$0.80
Output / 1M tokens$3.20
WeightsOpen
Parameters218B total / 25B active (sparse MoE)
Reasoning levelsNot verifiedUnverified
Modalitiestext, image
LicenseApache 2.0
API model idcommand-a-plus-05-2026
ReleasedMay 20, 2026

Pricing tiers: Azure's current USD retail feed lists Command A Plus public-cloud Global at $0.80/$3.20 per MTok and data-zone meters at $0.88/$3.52. US Government Global is $1.00/$4.00 and data-zone $1.10/$4.40. These are Azure rates, not Cohere-direct prices. Cohere's A+ page says trial and production keys are free until rate limits; general paid-key documentation and production-access requirements differ. Standard Vault lists A+ L/XL at $17.50/$32.50 per instance-hour, typically via a waitlist; encrypted beta design-partner access is separately scoped.

Publisher evidence

Published benchmark results

Each result keeps its source date visible. Dates refer to source publication or capture unless the source explicitly states when it measured the result. A missing benchmark is not treated as a zero.

Published benchmark results for Cohere Command A+
BenchmarkScoreSource dateSource
Artificial Analysis Intelligence IndexCommand A+132026-10-01Artificial Analysis · View source

Best for

  • Open-weight deployments
  • Enterprise RAG
  • Tool use

Watch out

The checked Cohere API ID has 128,000 context and 64,000 maximum output. Its W4A4 checkpoint config lists 200,000 positions; that does not expand the hosted limit. Thinking is enabled by default and can be disabled with thinking.type=disabled; thinking.token_budget is a budget rather than a low/medium/high menu. Cohere recommends leaving at least 1K tokens for the answer. Trial A+ is 20 requests/minute; the rate-limit page caps trial/newer-model keys at 1,000 calls/month and directs production access to sales. Do not treat the model-specific free-use statement as unlimited production capacity or as an Azure/Vault entitlement.

Common questions

Cohere Command A+

Answered from the verified figures on this page rather than general guidance.

How much does Cohere Command A+ cost per million tokens?
Cohere Command A+ lists base rates of $0.80 per million input tokens and $3.20 per million output tokens. Estimate both input and output token volume and any additional charges for your workload. Azure's current USD retail feed lists Command A Plus public-cloud Global at $0.80/$3.20 per MTok and data-zone meters at $0.88/$3.52. US Government Global is $1.00/$4.00 and data-zone $1.10/$4.40. These are Azure rates, not Cohere-direct prices. Cohere's A+ page says trial and production keys are free until rate limits; general paid-key documentation and production-access requirements differ. Standard Vault lists A+ L/XL at $17.50/$32.50 per instance-hour, typically via a waitlist; encrypted beta design-partner access is separately scoped.
What is Cohere Command A+'s context window?
Cohere Command A+ accepts about 128K tokens of context. That only matters if you routinely send very long documents, large codebases, or multi-turn histories that approach that limit.
What is Cohere Command A+ best for?
Cohere Command A+ is a balanced tier from Cohere. It suits open-weight deployments, enterprise rag, tool use. The checked Cohere API ID has 128,000 context and 64,000 maximum output. Its W4A4 checkpoint config lists 200,000 positions; that does not expand the hosted limit. Thinking is enabled by default and can be disabled with thinking.type=disabled; thinking.token_budget is a budget rather than a low/medium/high menu. Cohere recommends leaving at least 1K tokens for the answer. Trial A+ is 20 requests/minute; the rate-limit page caps trial/newer-model keys at 1,000 calls/month and directs production access to sales. Do not treat the model-specific free-use statement as unlimited production capacity or as an Azure/Vault entitlement.
Can I self-host Cohere Command A+?
Cohere Command A+ publishes open weights, but self-hosting depends on the licence, hardware footprint, quantisation quality, and serving stack. A hosted API is often cheaper until you have measured throughput and concurrency on your own hardware.