Skip to content
localmodel.run

Text model · Command

CR Command A: RAM and VRAM requirements

Command A needs about 70.4 GB to run at Q4_K_M (the Q4_K_M GGUF file is ~67.1 GB to download; KV cache and overhead add the rest), or about 121.3 GB at Q8_0. The lightest hardware that runs it is Apple M4 Max (128GB).

Command family · 111B params · released Mar 2025.

Shopping for hardware? See what runs Command A →

License cc-by-nc-4.0 · Non-commercial ↓ 1.8K/mo ♥ 394 on HuggingFace
Q4_K_M GGUF
67.1 GB
Q8_0 GGUF
118 GB
Memory @ Q4 (4k)
~70.4 GB
Context
256 k

Q4_K_M and Q8_0 are GGUF download (file) sizes. Memory @ Q4 includes the KV cache and overhead, so it runs larger than the file on disk.

Will it run on your device?

Command A runs on 4 of 40 tracked devices at Q4_K_M.

4 run well 0 tight fit 36 too small
Fit check Q4_K_M
No, not enough memory
needs 70.4 GB usable 10.5 GB
See the full breakdown

Memory breakdown

Weights (Q4_K_M)67.1 GB
+
KV cache (4k)2.5 GB
+
Overhead0.8 GB
=
Total70.4 GB

How context length changes it

4k context ~70.4 GB 32k context ~88.1 GB 128k context ~148.8 GB

Longer context grows the KV cache quickly: Command A needs ~70.4 GB at 4k but ~148.8 GB at 128k, which can push it past a device that fits it at a short context.

Speed drops too: every token re-reads the KV cache, so at 128k context Command A generates at roughly ~45% of its short-context speed. How this is estimated.

Benchmark scores

Aider polyglot · coding

#9 of 10 ranked · Percent of 225 Exercism exercises solved across six languages.

12%

BFCL · tool use

#3 of 19 ranked · Overall Acc on the Berkeley Function-Calling Leaderboard.

46.5%

Sourced third-party benchmarks for the full-precision Command A, not a specific quant. See the full leaderboard.

Quantization sizes

GGUF quantson disk
Quantization Size on disk
Q2_K 46.5 GB est
Q3_K_M 54.3 GB est
Q4_K_M (default) 67.1 GB
Q5_K_M 79.1 GB est
Q6_K 91 GB est
Q8_0 118 GB
FP16 222 GB

Lower quant = smaller and faster, slightly lower quality. Q4_K_M is the common default.

Run it

Ollama
$ ollama run command-a:111b
llama.cpp
$ llama-cli -hf lmstudio-community/c4ai-command-a-03-2025-GGUF:Q4_K_M
LM Studio
$ lms get lmstudio-community/c4ai-command-a-03-2025-GGUF

Which devices can run Command A?

Apple Silicon Macs

RAM-only laptops

iPhone & iPad

Android

NVIDIA GPUs

AMD GPUs

Same job, different size

Pick by what fits your memory: step down to free up VRAM, or step up if you have headroom.

Next size down Llama 4 Scout 109B Next size up gpt-oss 120B 117B

Similar models

FAQ

How much VRAM or RAM does Command A need?

At Q4_K_M, Command A needs about 70.4 GB (weights ~67.1 GB + KV cache + overhead) at a 4k context. At Q8_0 budget ~121.3 GB.

What is the Q4_K_M GGUF file size of Command A?

The Q4_K_M GGUF file is about 67.1 GB to download, and the Q8_0 GGUF is about 118 GB. That is the weights file on disk; to run it you also need room for the KV cache and overhead, so budget ~70.4 GB of memory at Q4_K_M.

Can Command A run on a laptop?

Command A is large; you need a high-memory Mac or multi-GPU setup at Q4_K_M.

Can I use Command A commercially?

No. cc-by-nc-4.0: non-commercial use only.

Understand the numbers

Short guides to the ideas behind Command A's memory and quant figures.

Cohere's 111B flagship with a 256K context, strong for RAG and tool use. Non-commercial license.

Sources

Last validated 2026-08-03. Memory figures are estimates. See methodology.