Cohere logo Open weights · CC-BY-NC

Command R7B

The smallest in Cohere's Command R series at 7B parameters, released December 2024, light enough for a single GPU with RAG and multilingual support. Weights under CC-BY-NC.

Jump to: Specs Pricing Self-hosting Similar models FAQ More from Cohere
Visit website Compare vs other LLMs →
List price via Cohere
Input
$0.04 /1M tokens
Output
$0.15 /1M tokens
Cohere 1 provider

Key Specifications

Context window
128K tokens
Max output
4K tokens
Released
Parameters
7B
Inputs
Text
Capabilities Show details
Function calling Connect to external tools, APIs, and systems. Structured output Return responses in structured formats like JSON.

Hosted API pricing

Provider Input /1M tokens Output /1M tokens Cost at 10M in + 2M out
Cohere logo Cohere Creator $0.04 $0.15 $0.70 View

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 R7B needs about 7 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 1× RTX 3060 at about $43 a month.

That costs the same as roughly 741M tokens a month on Cohere's API. Self-hosting is more expensive below that volume.

Fits on 1 Single node ≤ 8
Precision Memory Cheapest fit (32K context) Cost /mo Break-even vs API
4-bitINT4 / FP4
7 GB
Nvidia logo 1× RTX 3060
$43
741M tokens /mo
8-bitFP8 / INT8
10 GB
Nvidia logo 1× RTX 3060
$43
741M tokens /mo
16-bitFP16 / BF16
17 GB
Nvidia logo 2× RTX 3060
$86
1B tokens /mo

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. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Similarly priced models

The models nearest Command R7B by blended rate, each at its own cheapest provider.

Model Blended /1M vs Command R7B
Meta Llama 3.1 8B Meta $0.025 −57%
DeepSeek DeepSeek OCR-2 DeepSeek $0.03 −49%
Alibaba Cloud Qwen3.7-Flash Alibaba Cloud $0.0467 −20%
OpenAI GPT-OSS-20B OpenAI $0.0483 −17%
Cohere Command R7B This model Cohere $0.0583
Google Cloud Gemma 3 12B Instruct Google Cloud $0.0583 0%
Meta Llama 3.2 1B Meta $0.0583 0%
OpenAI GPT-OSS-120B OpenAI $0.0633 +9%
Nvidia Nemotron 3 Nano 30B A3B Nvidia $0.075 +29%
Meta Llama 3 8B Meta $0.0833 +43%
Alibaba Cloud Qwen3-Coder-30B-A3B Alibaba Cloud $0.0917 +57%

Prices are USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Frequently Asked Questions

What is Command R7B good for?

High-volume RAG, tool use and multilingual tasks. CC-BY-NC weights at the light end of Cohere's lineup, small enough for a single consumer GPU.

When is Command R7B not a good fit?

Image input: it is text only. Commercial self-hosting needs a separate licence from Cohere, and its short maximum output limits long answers.

What is the cheapest way to run Command R7B?

Hosted, unless you push serious volume. Cohere charges $0.04 in / $0.15 out per 1M tokens. The cheapest rental that fits is 1x RTX 3060 at $43 a month, which costs the same as about 741M tokens a month on that API.

Can I self-host Command R7B?

Yes. The weights are CC-BY-NC licensed. At 4-bit it needs about 7 GB of GPU memory, which starts at roughly $43 a month on the cheapest rental that fits.

More from Cohere

Model Context Input /1M Output /1M
Cohere Command R 08-2024 Open weights 128K $0.15 $0.60
Cohere Command A+ Open weights 128K $0.80 $3.20
Cohere Command R+ 08-2024 Open weights 128K $2.50 $10.00
Cohere Command A Open weights 256K $2.50 $10.00
Cohere Command A Reasoning Open weights 256K No hosted price
Cohere Command A Vision Open weights 128K No hosted price
Cohere Command A Translate Open weights 8K No hosted price