CohereCommand R

C4ai Command R V01 — Hardware Requirements & GPU Compatibility

ChatFunctions

C4AI Command R is a 35-billion-parameter instruction-tuned model released by Cohere For AI in March 2024. It was built for conversational use with a particular focus on retrieval-augmented generation and tool use, including grounded answers that cite the source documents they draw on, and it was trained to work across ten major languages. It supports a 128K-token context window and is released under a Creative Commons Attribution-NonCommercial 4.0 license, so commercial use is not permitted. At roughly 35 billion parameters, 4-bit quantization needs around 20GB of memory, within reach of a single 24GB consumer GPU.

35.0B params · denseCC BY-NC 4.0Mar 202467.0K downloads1.1K likes7.6K quant downloads

Your hardware

Verdicts below are for the reference RTX 4090. Set yours to personalise the page.

Detection reads your browser's GPU name with no data sent anywhere.

C

Usable on a 24 GB GPU

Runs from 16 GB (IQ2_S · 16K)Best on 24 GB (Q4_K_M · ~28 tok/s · 4K)Macs from 18 GB (IQ2_XXS · ~10 tok/s)

CC BY-NC 4.0

Reference GPU · RTX 4090 24 GB

Estimates, ±20% · how we estimate

Run C4ai Command R V01

One recommended way. Every other method, OS and quant is in the step-by-step guide.

Recommended

Ollama

one command · needs Ollama installed

$ollama run hf.co/mradermacher/c4ai-command-r-v01-i1-GGUF:Q4_K_M

Q4_K_M needs offload on the RTX 4090 · 3.0 GB short · 16K context

Download: c4ai-command-r-v01.i1-Q4_K_M.gguf21.53 GBmradermacher

install, OS, other quants

Other ways

each opens step-by-step

Run in cloud

Fits on A40 (48 GB) (21 GB headroom) · Q4_K_M

Generation speed: ~20 tok/s; generation speed
Cost per 1M output tokens: $4.96; per 1M output tokens

Deployon RunPodBest optionfrom $0.35/hr

Compare GPUs →

or

How Much VRAM Does C4ai Command R V01 Need?

Select a quantization level; the listed hardware adheres to it.

Quantization	VRAM	+32K	+128K	File	Quality	On RTX 4090 (reference)
	10.6 GB	18.9 GB	45.5 GB	10.2 GB	2-bit · i-quant extreme compression — Importance-weighted 2-bit, extreme compression — significant quality loss	Fits: 32K context · ~62 tok/s
	11.5 GB	19.9 GB	46.5 GB	11.1 GB	2-bit · i-quant extra small — Importance-weighted 2-bit, extra small	Fits: 32K context · ~57 tok/s
	12.0 GB	20.3 GB	46.9 GB	11.8 GB	2-bit · i-quant small — Importance-weighted 2-bit, small	Fits: 32K context · ~55 tok/s
	13.0 GB	21.3 GB	47.9 GB	12.7 GB	2-bit · i-quant medium — Importance-weighted 2-bit, medium	Fits: 32K context · ~50 tok/s
	14.9 GB	23.2 GB	49.8 GB	13.8 GB	3-bit · i-quant — Importance-weighted 3-bit	Fits: 32K context · ~44 tok/s
	15.4 GB	23.7 GB	50.3 GB	12.7 GB	2-bit · small K-quant — 2-bit small K-quant	Fits: 32K context · ~43 tok/s
	15.9 GB	24.2 GB	50.8 GB	15.1 GB	3-bit · i-quant extra small — Importance-weighted 3-bit, extra small	Fits: 16K context · ~41 tok/s
	16.4 GB	24.7 GB	51.3 GB	15.9 GB	3-bit · i-quant small — Importance-weighted 3-bit, small	Fits: 16K context · ~40 tok/s
	16.4 GB	24.7 GB	51.3 GB	13.8 GB	2-bit · K-quant — 2-bit quantization with K-quant improvements	Fits: 16K context · ~40 tok/s
	16.8 GB	25.1 GB	51.7 GB	15.9 GB	3-bit small — 3-bit small quantization	Fits: 16K context · ~39 tok/s
	17.3 GB	25.6 GB	52.2 GB	16.7 GB	3-bit · i-quant medium — Importance-weighted 3-bit, medium	Fits: 16K context · ~38 tok/s
	18.8 GB	27.1 GB	53.7 GB	17.6 GB	3-bit medium — 3-bit medium quantization	Fits: 16K context · ~35 tok/s
	19.2 GB	27.6 GB	54.2 GB	20.3 GB	4-bit · legacy — 4-bit legacy quantization	Fits: 16K context · ~34 tok/s
	19.7 GB	28 GB	54.6 GB	19.1 GB	3-bit large — 3-bit large quantization	Fits: 16K context · ~33 tok/s
	20.7 GB	29 GB	55.6 GB	19.2 GB	4-bit · i-quant compact — Importance-weighted 4-bit, compact	Fits: 8K context · ~32 tok/s
	21.6 GB	30 GB	56.6 GB	20.4 GB	4-bit small — 4-bit small quantization	Fits: 8K context · ~30 tok/s
	23.1 GB	31.4 GB	58 GB	21.5 GB	4-bit medium — 4-bit medium quantization — most popular sweet spot	Fits: 4K context · ~28 tok/s
	24.1 GB	32.4 GB	59 GB	24.3 GB	5-bit · legacy — 5-bit legacy quantization	Does not fit: needs 27.9 GB
	26.4 GB	34.8 GB	61.4 GB	24.3 GB	5-bit small — 5-bit small quantization	Does not fit: needs 30.3 GB
	27.4 GB	35.7 GB	62.3 GB	25 GB	5-bit medium — 5-bit medium quantization — good quality/size tradeoff	Does not fit: needs 31.3 GB
	31.8 GB	40.1 GB	66.7 GB	28.7 GB	6-bit, very good — 6-bit quantization, very good quality	Does not fit: needs 35.6 GB
	38.5 GB	46.8 GB	73.4 GB	37.2 GB	8-bit, near-lossless — 8-bit quantization, near-lossless	Does not fit: needs 42.4 GB
	77.0 GB	85.3 GB	111.9 GB	~70 GB	16-bit · bfloat — Brain floating point 16 — preferred for training	Does not fit: needs 80.8 GB

est.calculated, no published file yet

Selected Q4_K_M

4-bit medium quantization

27.0GBat 16K context

Weights: 22.2 GB
KV cache: 4.4 GB
Overhead: 0.3 GB

Fits RTX 4090 (reference) · fits up to 4K context · ~28 tok/s

≥50 tok/s needs ≥1.77 TB/s · RTX 5090 ~50

Or rent A40 from $0.35/hr.

Which Hardware Can Run C4ai Command R V01?

Q4_K_M · 23.1 GB

C4ai Command R V01 (Q4_K_M) requires 23.1 GB of VRAM to load the model weights. For comfortable inference with headroom for KV cache and system overhead, 31+ GB is recommended. 7 GPUs can run it, including NVIDIA GeForce RTX 5090. 100 devices with unified memory can run C4ai Command R V01, including Mac Studio (M3 Ultra, 256GB), Mac Studio (M3 Ultra, 512GB), Mac Studio M5 Max (36 GB).

By VRAM tier · typical GPUs and Macs at Q4_K_M

Typical hardwareQuantVRAMtok/sContext that fits

8–12 GB · RTX 4060, RTX 3060 12GBDoesn't fit — smallest quant needs 14.5 GB8 GB tier →12 GB tier →

16 GB · RTX 5060 Ti 16GBIQ2_S · 15.9 GB · ~24 tok/s · 16K15.9~2416K16 GB tier →

24 GB · RTX 4090referenceQ4_K_M · 27 GB · ~28 tok/s · 4Ktight27~284K24 GB tier →

Mac 32 GB · Mac Mini M4Q4_K_M · 27 GB · ~4 tok/s · 16K27~416KMacs →

Mac 64 GB · Mac Studio M4 MaxQ4_K_M · 27 GB · ~17 tok/s · 32K27~1732KMacs →

How good is C4ai Command R V01?

llmrun Score blends public benchmark boards into one number; 100 = the average open model.

Not enough data yet for an llmrun Score.

A Score needs results on enough independent boards. How the Composite Score is built

Benchmarks

LMArena Text
LMArena · Oct 2026Arena
1227.0
#136/187 open
Arena
MMLU-Pro
TIGER-LabIndependent
37.9%
#108/145 open
Independent

All 2 C4ai Command R V01 benchmarks

Alternatives to C4ai Command R V01

Command R family · 5 models →

Similar models

6tok/s on RTX 4090

Model	Score	VRAM	tok/s	Why it is similar
MiniCPM V 4.7 35B A3BOpenBMB	—	22.1 GB	~169	same size
Llm Jp 4.1 33B Thinkingllm-jp	—	22.5 GB	~35	same size
Agnes 3.0 FlashAgnes-AI	—	22.1 GB	~35	same size
Ornith 1.5 35B A3Bornith-ai	—	22 GB	~151	same size
Agents A1InternScience	—	21.5 GB	~171	same size
Intern S2 PreviewInternLM	—	73.2 GB	—needs > 24 GB	same size

Specifications

Publisher: Cohere
Family: Command R
Parameters: 35.0B
Release Date: 2024-03-11
License: CC BY-NC 4.0

Resources

Frequently Asked Questions

How much VRAM does C4ai Command R V01 need?

C4ai Command R V01 requires 23.1 GB of VRAM at Q4_K_M, or 77.0 GB at BF16.

VRAM = Weights + KV Cache + Overhead

Weights = 35.0B × 4.8 bits ÷ 8 = 21 GB

KV Cache + Overhead ≈ 2.1 GB (at 2K context + ~0.3 GB framework)

Fit ratings and hardware model lists check this model with room for a 16K-token context, which needs a little more memory.

VRAM usage by quantization

Q4_K_M

23.1 GB

Learn more about VRAM estimation →

Can NVIDIA GeForce RTX 4090 run C4ai Command R V01?

Yes, at Q4_K_M (23.1 GB) or lower. Higher quantizations like Q5_0 (24.1 GB) exceed the NVIDIA GeForce RTX 4090's 24 GB.

What's the best quantization for C4ai Command R V01?

For C4ai Command R V01, Q4_K_M (23.1 GB) offers the best balance of quality and VRAM usage. Q5_0 (24.1 GB) provides better quality if you have the VRAM. The smallest option is IQ2_XXS at 10.6 GB.

VRAM requirement by quantization

IQ2_XXS

10.6 GB~53%

IQ3_XS

15.9 GB~73%

Q3_K_M

18.8 GB~83%

Q4_K_M ★

23.1 GB~89%

Q5_0

24.1 GB~90%

BF16

77.0 GB~100%

★ Recommended — best balance of quality and VRAM usage.

Learn more about quantization →

Can I run C4ai Command R V01 on a Mac?

Yes — MacBook Pro 14" M3 Pro (18 GB) and 108 other Macs can run C4ai Command R V01, the smallest at IQ2_XXS. Apple Silicon uses unified memory, so the model shares RAM with the system. At Q4_K_M you need at least 23.1 GB of usable unified memory (RAM minus macOS overhead), from Mac Studio M1 Max (32 GB).

Can I run C4ai Command R V01 locally?

Yes — C4ai Command R V01 can run locally on consumer hardware. Popular tools include Ollama, LM Studio, and llama.cpp.

How fast is C4ai Command R V01?

At Q4_K_M, C4ai Command R V01 can reach ~208 tok/s on AMD Instinct MI350X. On NVIDIA GeForce RTX 4090: ~28 tok/s. Speed depends mainly on GPU memory bandwidth. Real-world results typically within ±20%.

tok/s = (bandwidth GB/s ÷ model GB) × efficiency

Example: NVIDIA B200 → 8000 ÷ 23.1 × 0.65 = ~225 tok/s

Estimated speed at Q4_K_M (23.1 GB)

NVIDIA B200

~225 tok/s

NVIDIA GeForce RTX 4090

~28 tok/s

NVIDIA B300

~225 tok/s

AMD Instinct MI350X

~208 tok/s

Real-world results typically within ±20%. Speed depends on batch size, quantization kernel, and software stack.

Learn more about tok/s estimation →

What's the download size of C4ai Command R V01?

At Q4_K_M, the download is about 20.99 GB. The full-precision BF16 version is 69.96 GB. The smallest option (IQ2_XXS) is 9.62 GB.

Which GPUs can run C4ai Command R V01?

7 consumer GPUs can run C4ai Command R V01 at Q4_K_M. Top options include AMD Radeon RX 7900 XTX, NVIDIA GeForce RTX 3090.

Which devices can run C4ai Command R V01?

100 devices with unified memory can run C4ai Command R V01 at Q4_K_M (23.1 GB), including AMD Ryzen AI 9 HX 370 (Strix Point) Laptop, ASUS Ascent GX10, Beelink GTR9 Pro (Ryzen AI Max+ 395, 128 GB), Framework Desktop (Ryzen AI Max+ 395, 128 GB). Apple Silicon Macs use unified memory shared between CPU and GPU, making them well-suited for local LLM inference.

Run C4ai Command R V01

27.0 GB VRAM at 16K context · needs offload on the RTX 4090 (3.0 GB short)

Works on macOS, Windows and Linux.

Install Ollama (Linux)
$curl -fsSL https://ollama.com/install.sh | sh
Install Ollama (macOS / Windows)
Download and run the app; it installs the ollama command.
Download Ollama ↗
Run Q4_K_M
$ollama run hf.co/mradermacher/c4ai-command-r-v01-i1-GGUF:Q4_K_M
Raise the context window
Type it inside the chat. For the API server instead, start it with OLLAMA_CONTEXT_LENGTH=16384 ollama serve.
$/set parameter num_ctx 16384
A larger context needs more memory than the weights alone.
Use the API
Ollama serves its API on http://localhost:11434.

Works on macOS, Windows and Linux.

Install LM Studio
Desktop app for macOS (Apple silicon), Windows and Linux.
lmstudio.ai ↗
Download the model
In the app press Ctrl/Cmd + Shift + M and paste https://huggingface.co/mradermacher/c4ai-command-r-v01-i1-GGUF, then pick Q4_K_M. Or use the CLI (run LM Studio once first):
$lms get https://huggingface.co/mradermacher/c4ai-command-r-v01-i1-GGUF@Q4_K_M
Open on Hugging Face ↗
Load and chat
Open the model loader, select the model and set Context Length to 16384. Leave GPU offload at maximum if it fits.
Start the local server (optional)
$lms server start

Works on macOS, Windows and Linux.

Install llama.cpp (macOS)
MacPorts, Nix and conda-forge also work.
$brew install llama.cpp
Install options ↗
Install llama.cpp (Windows)
$winget install llama.cpp
Install llama.cpp (Linux)
Download the prebuilt llama-*-bin-ubuntu-* archive for your GPU backend (CPU, Vulkan) from the releases page, unpack it and run the binaries from that folder. Homebrew (brew install llama.cpp) and conda-forge also work on Linux.
llama.cpp releases ↗
Start the server
Downloads the model on first run. The chat UI and an OpenAI-compatible API are at http://localhost:8080.
$llama-server -hf mradermacher/c4ai-command-r-v01-i1-GGUF:Q4_K_M -c 16384 -ngl 99 --port 8080

Works on macOS, Windows and Linux.

Option A: Docker Model Runner
Docker Desktop: Settings > AI > Enable Docker Model Runner. Docker Engine on Linux: install the docker-model-plugin package.
$docker model run hf.co/mradermacher/c4ai-command-r-v01-i1-GGUF:Q4_K_M
Model Runner docs ↗
The quant tag after the repo is how Hugging Face cards write it; if your Docker version rejects it, drop the tag and pick the quant by repo.
Option B: llama.cpp server image (NVIDIA)
Needs the NVIDIA Container Toolkit. Models are cached in the llama-cache volume. Open http://localhost:8080.
$docker run --gpus all -p 8080:8080 -v llama-cache:/root/.cache ghcr.io/ggml-org/llama.cpp:server-cuda -hf mradermacher/c4ai-command-r-v01-i1-GGUF:Q4_K_M -c 16384 -ngl 99 --host 0.0.0.0 --port 8080
Image tags (cuda13, vulkan, rocm) ↗
Cache path inside the image is not in the docs; remove the -v flag if the volume complains.

Linux only.

Install vLLM
Linux only; on Windows run it inside WSL2. NVIDIA GPU shown; AMD ROCm, Intel and CPU builds have their own install pages.
$uv venv --python 3.12 --seed
$source .venv/bin/activate
$uv pip install vllm --torch-backend=auto
vLLM installation ↗
Serve the model
OpenAI-compatible API at http://localhost:8000/v1.
$vllm serve CohereLabs/c4ai-command-r-v01 --max-model-len 16384
Runs the original weights, not the Q4_K_M GGUF; needs ~77 GB of GPU memory.

Works on macOS, Windows and Linux.

Install Jan
jan.ai ↗
Find the model
Open the Hub tab, paste this model ID in Search, and choose Q4_K_M if offered.
$mradermacher/c4ai-command-r-v01-i1-GGUF
Use the local API (optional)
Settings > Local API Server > Start Server. It listens on http://127.0.0.1:1337.

Rent GPUs hourly to run Q4_K_M right there. Pricing comes from partner marketplaces and varies.

Available GPU rentals for Q4_K_M, prioritized by value
GPU	Speed	$/1M tokens	Rent
A40 (48 GB)CheapestCheapest21 GB free	~20 tok/s	$4.96	Deploy on RunPodfrom $0.35/hr (opens in new tab)	48 GB · 21 GB free · ~20 tok/s · $4.96 / 1M tokens
A40 (48 GB)21 GB free	~20 tok/s	$5.71	Deploy on Vast.aifrom $0.40/hr (opens in new tab)	48 GB · 21 GB free · ~20 tok/s · $5.71 / 1M tokens
RTX 6000 Ada (48 GB)21 GB free	~27 tok/s	$7.55	Deploy on Vast.aifrom $0.73/hr (opens in new tab)	48 GB · 21 GB free · ~27 tok/s · $7.55 / 1M tokens
RTX 6000 Ada (48 GB)21 GB free	~27 tok/s	$7.61	Deploy on RunPodfrom $0.74/hr (opens in new tab)	48 GB · 21 GB free · ~27 tok/s · $7.61 / 1M tokens
L40S (48 GB)21 GB free	~24 tok/s	$9.03	Deploy on RunPodfrom $0.79/hr (opens in new tab)	48 GB · 21 GB free · ~24 tok/s · $9.03 / 1M tokens
L40S (48 GB)21 GB free	~24 tok/s	$9.16	Deploy on Vast.aifrom $0.80/hr (opens in new tab)	48 GB · 21 GB free · ~24 tok/s · $9.16 / 1M tokens
A100 80GB53 GB free	~57 tok/s	$6.73	Deploy on RunPodfrom $1.39/hr (opens in new tab)	53 GB free · ~57 tok/s · $6.73 / 1M tokens
RTX PRO 6000 (96 GB)69 GB free	~45 tok/s	$9.48	Deploy on Vast.aifrom $1.54/hr (opens in new tab)	96 GB · 69 GB free · ~45 tok/s · $9.48 / 1M tokens

Compare GPUs →