Command R Models — Hardware Requirements
5 Command R models from Cohere and the community, from the smallest that runs in 7.9 GB of VRAM at Q4 up to 103.8B parameters. Every row links to full quantization tables and GPU compatibility.
All Command R Models by Size
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| C4ai Command R 08 2024 | 32.3B | 25.1 GB | — | ||
| C4ai Command R V01 | 35.0B | 27.0 GB | — | ||
| C4ai Command R Plus 08 2024 | 103.8B | 73.8 GB | — | ||
| C4ai Command R Plus | 103.8B | 73.8 GB | — |
Community variants and fine-tunes(1)
| Model | Params | Runs from (Q4) | Context | Publisher | Quant downloads |
|---|---|---|---|---|---|
| C4ai Command R7b 12 2024 | 8.0B | 7.9 GB | — |
Frequently Asked Questions
- How much VRAM do I need to run a Command R model?
- The smallest Command R model, C4ai Command R7b 12 2024, runs from 7.9 GB of VRAM at Q4_K_M (smaller low-bit quantizations start at 3.0 GB with a noticeable quality loss). Larger family members need proportionally more — see the table above for every model.
- Which Command R models can I run on a 16 GB GPU?
- 1 of 5 Command R models fit in 16 GB of VRAM at Q4_K_M, including C4ai Command R7b 12 2024.
- What is the most popular Command R model to run locally?
- C4ai Command R7b 12 2024 is the most downloaded Command R model in local-friendly quantized formats. It runs from 7.9 GB of VRAM at Q4_K_M.