We may earn a commission if you buy through links on this page. It never changes a score or a ranking. How we make money
There's a free plan; paid tiers are quoted.
Cohere’s Trial API key offers free calls for prototyping, but it is rate-limited and cannot be used for production or commercial work. A Production API key requires production access and is billed pay as you go, at month end or when $250 is outstanding. North and Compass have custom enterprise pricing. Model Vault charges by GPU-hour, with rates depending on the GPU configuration and optional confidential compute.
01Every plan, line by line
| Plan | Billed annually | Billed monthly | Built for |
|---|---|---|---|
| Trial API key | — | Quote | Prototyping with Cohere APIs |
| Production API key | — | Quote | Production API workloads |
| North | — | Quote | Enterprise AI teams |
| Compass | — | Quote | Business search and discovery |
| Model Vault — L4 24GB | — | Quote | Managed model inference |
| Model Vault — A100 40GB | — | Quote | Managed model inference |
| Model Vault — A100 80GB | — | Quote | Managed model inference |
| Model Vault — MI300X 192GB | — | Quote | Managed model inference |
| Model Vault — H100 80GB | — | Quote | Managed model inference |
| Model Vault — H200 141GB | — | Quote | Managed model inference |
| Model Vault — H100 80GB encrypted | — | Quote | Confidential-compute inference |
| Model Vault — B200 192GB | — | Quote | Managed model inference |
| Model Vault — B200 192GB encrypted | — | Quote | Confidential-compute inference |
| Model Vault — GB200 192GB | — | Quote | Managed model inference |
Add-ons and overages
- Model Vault reserved capacity is available; contact sales for pricing.
- Further promotional pricing is available for monthly Model Vault commits.
- Legacy API rates for existing customers: Command $1.00/1M input tokens and $2.00/1M output; Command-light $0.30/$0.60; Command R 03-2024 $0.50/$1.50; Command R+ 04-2024 $3.00/$15.00; Command R+ 08-2024 $2.50/$10.00.
- Aya Expanse 8B and 32B API rates: $0.50/1M input tokens and $1.50/1M output tokens.
02What the plan grid doesn't tell you
Production API usage
Production is pay as you go. Ask Cohere for current model-specific rates and estimate costs against expected input and output usage before launch.
Trial rate limits
Trial calls are free but rate-limited. The pricing information does not give a numeric limit; production use requires a Production API key.
GPU hours and commits
Model Vault bills per GPU-hour. Reserved capacity and further promotional pricing for monthly commits require contacting sales.
Enterprise product quotes
North and Compass have custom enterprise pricing, so teams need a Cohere quote to budget those products.
03Which plan should you buy?
| If you are… | Get | Why |
|---|---|---|
| Prototype or evaluation | Trial API key | Free API calls suit prototyping, but rate limits apply and production or commercial use is not allowed. |
| Production API workload | Production API key | It enables production access and pay-as-you-go billing; request model-specific rates before estimating costs. |
| Private inference deployment | Model Vault — L4 24GB | The listed on-demand rate is $3.00 per GPU-hour, or $1.00 through March 31, 2027. |
04Is Cohere worth the money?
Cohere’s free tier is useful for testing API calls, but it is not a production allowance: trial keys are rate-limited and barred from commercial use. For production APIs, get a quote for the models and volumes you plan to run because current model-specific rates are not published here. Model Vault has visible GPU-hour prices, which helps with compute estimates, but sustained workloads, reserved capacity and North or Compass require a closer sales discussion. Compare the billed workload against the private deployment requirements before choosing.
Cohere pricing FAQ
Is Cohere free to use?
Cohere offers free calls with a Trial API key. The key is rate-limited and cannot be used for production or commercial workloads. Production API access requires a Production API key and pay-as-you-go billing.
How much does Cohere’s production API cost?
Production API access is billed pay as you go, at month end or when $250 is outstanding. Current model-specific rates are not published in this pricing information, so request a quote or rate details for your intended models and usage.
What does Cohere Model Vault cost?
Model Vault is priced per GPU-hour. Listed on-demand prices start at $3.00 per hour for L4 24GB and rise by GPU configuration. Promotional rates are listed through March 31, 2027; reserved capacity requires contacting sales.
Does Cohere offer promotional Model Vault pricing?
Yes. Promotional per-GPU-hour rates are listed through March 31, 2027. The pricing information also notes further promotional pricing for monthly Model Vault commits, with details available from Cohere sales.