Models / Cohere/ Command A

Command A

Cohere · released Mar 11, 2025 · CohereForAI/c4ai-command-a-03-2025

Input: text. Output: text.InputOutput
Type
Open weightsCreative Commons Attribution-NonCommercial 4.0
Params
111B
Context
256K

about 192K words of context · download allowed, licence restricts use

Our take

Written Aug 3, 2026

Command A is a 111.1-billion-parameter text model from Cohere, released in March 2025 under a non-commercial licence. It is positioned as a research and evaluation asset rather than a production-ready option, with no benchmark scores in our data.

Who should pick it

Pick this for academic research or non-commercial exploration where you need to evaluate Cohere's architecture at large scale. Use it to test document-scale tasks with its 256,000-token request limit before committing to a commercial API contract. Skip it if you need a permissive licence for commercial products, if you require measured quality scores, or if you want to self-host for business use.

The case for it

  • 111.1 billion total parameters, a substantial scale among openly available weights.
  • 256,000-token request limit, enough for long documents.
  • Identical pricing across both tracked providers, so there is no arbitrage friction.

The case against it

  • Non-commercial licence only: no commercial products, no closed derivatives, and most fine-tuning redistribution is blocked.
  • No benchmark scores in our data — no measured Elo, reasoning, coding or general quality figures.
  • Throughput data is thin: only Cohere-hosted speed is known, and nothing we hold verifies performance elsewhere.
00

How good is it?

We hold no score for this model.

We look for every model we track on every board we watch, and none of them has turned up Command A — so there is no intelligence, coding, agentic or writing score to show you, not a low one, none.

The boards we watch →

01

Can you run it yourself?

Fits in memory
weights load entirely on the card
Spills to system RAM
some weights offload; much slower
Too large
will not load even with offload
est
size is calculated; the verdict could change by 10%
A card many people ownToo large

GeForce RTX 4090 · 24 GB

Weights at Q4_K_M70.1 / 24 GBest
Usable contextNot calculated for spilled setups
Decode speedNot estimated for spilled setups

Too large for this card. The weights do not fit even with part of them offloaded to system memory.

One step upToo large

Apple M1 Pro (16-core GPU) · 32 GB

Weights at Q4_K_M70.1 / 32 GBest
Usable contextNot calculated for spilled setups
Decode speedNot estimated for spilled setups

Too large for this card. The weights do not fit even with part of them offloaded to system memory.

Comfortable fit

On a MacFits in memory

Apple M1 Ultra (64-core GPU) · 128 GB

Weights at Q4_K_M70.1 / 128 GBest
Spare memory22.3 GB spare
Usable context66K of 256K
Decode speed8 tok/sest

Room to spare. 22.3 GB spare means a 10% error in the size would not change the answer.

Your hardware
Checking your profile…

Memory use by level

Against a 24 GB card.

Q4_K_M
recommended
70.1 GBest
Too large
Q5_K_M
82.2 GBest
Too large
Q8_0
122.8 GBest
Too large

All 3 sizes here are calculated, not measured. We hold no measured file for this model, so each size comes from the parameter count and every verdict above inherits that uncertainty.

Check against your own machine → · All 71 devices, with every size →

02

Or rent it from someone else

Cheapest published offer

Cheapest of 2 live listings. Picked at the widest standard context we hold, within one quantisation slice, so the numbers beside it are a price one host actually charges.

per 1M tokens
$2.50 in / $10.00 out
Context served
256K
Throughput
Not measured
Current provider offers with price, context and prompt-privacy answers
ProviderIn / out per 1M tokensContextThroughputTrains on promptsLogs promptsZero retention
OpenRouter$2.50 / $10.00256Knot measuredUnknownUnknownUnknown
Cohere$2.50 / $10.00256K59 tok/sNoYes30 daysUnknown

Across the 2 listings we hold: 1 say they do not train on prompts, 0 say they do and 1 do not say. 0 appear in the zero-retention registry we check; the rest are unknown to us rather than confirmed either way.

What each host's API supports

From the parameter list each endpoint publishes. Streaming is omitted: nothing we hold reports it, for any model.

Supported
Not supported
Not published
host gave no parameter list
API features per host
ProviderTool callingJSON outputStrict schema
OpenRouter
Cohere

Tool calling: 0 of 2 listings say yes, 2 say no. JSON output: 2 of 2 listings say yes. Strict schema: 2 of 2 listings say yes.

03

When we formed this view

Dates behind this page

Jul 26, 2026ListedListed on LLMapfirst indexed by our pipeline
Mar 11, 2025AnnouncedCommand A announced by Cohere

Prices last checked 6h ago

What we do not know about this model yet

  • We hold no measured file for it, so all 3 sizes on this page are calculated from the parameter count.
  • No board we watch has turned up a score, so we hold no quality figures at all.
  • Nothing we hold says whether an endpoint streams, so we do not show it either way.
  • 1 of 2 listings do not say whether they train on prompts.
  • We hold no cached-input rate for any of its listings.
04

Licence and identifiers

What the licence allowsCreative Commons Attribution-NonCommercial 4.0, what it allows commercially, and the identifiers you need to pull this model — its Hugging Face repo, our slug and a machine-readable card.

Licence

Creative Commons Attribution-NonCommercial 4.0

restricted_openNon-commercial

Weights are downloadable but commercial use is prohibited. Research and personal use only.

Identifiers

Modality record
text->text
Catalogue slug
cohere-command-a

Machine-readable model card (omc.json) →

Something wrong on this page? Tell us