Can Codestral 2 25.08 run on RTX 4000 Ada 20GB?

YES — Tight Fit

A83Great

Estimated from fit model

Codestral 2 25.08 needs ~18.8 GB VRAM. RTX 4000 Ada 20GB has 20.0 GB. With Q4_K_M quantization, expect ~20 tok/s.

Runtime: llama.cppCapacity: TightBandwidth: LowStack: StandardBottleneck: Balanced

Operating mode

Choose the run profile you care about

Interactive favors responsiveness, while light API and scale-out lean harder on serving readiness. The fit stays the same, but the recommendation lens changes.

Current mode

Balanced

Balanced for general local use. Keeps the ranking neutral across personal and serving workflows.

Capabilities:

Select quantization to explore

Q4_K_M (Medium quality) — 18.8 GB, 20.1 tok/s, Tight fit

18.8 GB required20.0 GB available

94% VRAM used

Fit status

Tight fit

Decode

20.1 tok/s

TTFT

9638 ms

Safe context

24K

Memory

18.8 GB / 20.0 GB

Memory breakdown

Weights13.4 GB

KV Cache2.4 GB

Runtime0.9 GB

Headroom2.0 GB

See how fast it feels

See how fast it feelsCodestral 2 25.08 on RTX 4000 Ada 20GB
1st promptCold start — includes initialization
>Write a Python function to sort a list
Here's a Python function that sorts a list using quicksort:

def quicksort(arr):
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

# Example usage
numbers = [3, 6, 8, 10, 1, 2, 1]
print(quicksort(numbers))  # [1, 1, 2, 3, 6, 8, 10]

The algorithm picks a pivot, partitions the list into smaller, equal, and larger elements, then recursively sorts each partition and concatenates the results.
2nd promptWarm — model ready, no init overhead
>Now add type hints and a docstring
def quicksort(arr: list[int]) -> list[int]:
    """Sort a list using the quicksort algorithm.

    Args:
        arr: List of integers to sort.

    Returns:
        A new sorted list.
    """
    if len(arr) <= 1:
        return arr
    pivot = arr[len(arr) // 2]
    left   = [x for x in arr if x < pivot]
    middle = [x for x in arr if x == pivot]
    right  = [x for x in arr if x > pivot]
    return quicksort(left) + middle + quicksort(right)

Added type hints for the input and return type, plus a Google-style docstring.
Estimated: 20.1 tok/s decode · 9.6s TTFT (warm) · 50 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Very little memory headroom

You can run the model, but there is not much room left for longer context, bigger batches, extra apps, or future model updates.

Best improvement path

Buy headroom, not only minimum fit

A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Tight fit	20.1 tok/s	5257 ms	24K
Coding	A	Tight fit	20.1 tok/s	9638 ms	24K
Agentic Coding	A	Runs with offload (needs ~0.8 GB host RAM)	13.3 tok/s	21138 ms	24K
Reasoning	A	Tight fit	20.1 tok/s	11391 ms	24K
RAG	A	Runs with offload (needs ~0.8 GB host RAM)	13.3 tok/s	26422 ms	24K

Inference speed

Codestral 2 25.08 inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for Codestral 2 25.08 at Q4_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~96 tok/s. Speed is memory-bandwidth bound, so cards that fit the whole model in VRAM run far faster than ones that offload to system RAM.

GPU / Mac	Memory	Quant	Speed (tok/s)	Fits?
RTX 5090 32GB	32 GB	Q4_K_M	96.2	Fits
RX 7900 XTX 24GB	24 GB	Q4_K_M	52.0	Fits
Mac Studio M3 Ultra 256GB	256 GB	Q4_K_M	41.9	Fits
RTX 4090 24GB	24 GB	Q4_K_M	41.7	Fits
RTX 3090 24GB	24 GB	Q4_K_M	38.2	Fits
MacBook Pro M4 Max 128GB	128 GB	Q4_K_M	35.2	Fits
MacBook Pro M4 Max 64GB	64 GB	Q4_K_M	35.2	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q4_K_M	34.9	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q4_K_M	33.1	Fits
MacBook Pro M4 Pro 48GB	48 GB	Q4_K_M	22.2	Fits
RTX 4080 Super 16GB	16 GB	Q4_K_M	18.6	Heavy offload
MacBook Pro M3 Max 64GB	64 GB	Q4_K_M	18.1	Fits
MacBook Pro M1 Max 64GB	64 GB	Q4_K_M	16.6	Fits
RTX 4070 12GB	12 GB	Q4_K_M	6.6	Too big
RTX 3060 12GB	12 GB	Q4_K_M	4.4	Too big
RTX 4060 8GB	8 GB	Q4_K_M	2.0	Too big

Estimates for single-stream decoding at Q4_K_M; real tokens/sec varies with prompt length, context, batch size, and runtime build. Prompt processing (prefill) is faster than the decode figures shown here.

Quantization options

How Codestral 2 25.08 (22B params) fits at each quantization level on RTX 4000 Ada 20GB (20.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	8.6 GB	Low	A84
Q3_K_S	3	10.8 GB	Low	A85
NVFP4	4	12.3 GB	Medium	A85
Q4_K_MBest for your GPU	4	13.4 GB	Medium	A84
Q5_K_M	5	15.8 GB	High	F0
Q6_K	6	18.0 GB	High	F0
Q8_0	8	23.5 GB	Very High	F0
F16	16	45.1 GB	Maximum	F0

Get started

Copy-paste commands to run Codestral 2 25.08 on your machine.

Run

lms load codestral-2508 && lms server start

Your hardware

More models your RTX 4000 Ada 20GB can run

Model	Params	Grade	Decode
Qwen3-Coder 30B A3B Instruct	30.5B	A	23.8 tok/s
Qwen 3.5 27B	27B	A	10.7 tok/s
Qwen 3.6 27B	27B	S	10.1 tok/s
Qwen3-VL 30B A3B Instruct	30B	A	25.3 tok/s
Magistral Small 2507	24B	S	20.6 tok/s

Frequently asked questions

Can RTX 4000 Ada 20GB run Codestral 2 25.08?

Yes, RTX 4000 Ada 20GB can run Codestral 2 25.08 with a A grade (Tight fit). Expected decode speed: 20.1 tok/s.

How much VRAM does Codestral 2 25.08 need?

Codestral 2 25.08 (22B parameters) requires approximately 18.8 GB of memory with Q4_K_M quantization.

What is the best quantization for Codestral 2 25.08?

The recommended quantization for Codestral 2 25.08 is Q4_K_M, which balances quality and memory efficiency.

What speed will Codestral 2 25.08 run at on RTX 4000 Ada 20GB?

On RTX 4000 Ada 20GB, Codestral 2 25.08 achieves approximately 20.1 tokens per second decode speed with a time-to-first-token of 9638ms using Q4_K_M quantization.

Can RTX 4000 Ada 20GB run Codestral 2 25.08 for coding?

For coding workloads, Codestral 2 25.08 on RTX 4000 Ada 20GB receives a A grade with 20.1 tok/s and 24K context.

What context window can Codestral 2 25.08 use on RTX 4000 Ada 20GB?

On RTX 4000 Ada 20GB, Codestral 2 25.08 can safely use up to 24K tokens of context. The model's official context limit is 256K, but available memory constrains the safe maximum.

What should I upgrade first if Codestral 2 25.08 feels slow on RTX 4000 Ada 20GB?

Buy headroom, not only minimum fit. A slightly larger memory tier gives you safer context growth and makes the recommendation more future-proof.

See all results for RTX 4000 Ada 20GB See all hardware for Codestral 2 25.08

Embed this result▼

Paste this snippet into any page to show a live fit card.

<iframe src="https://willitrunai.com/embed/codestral-2-25.08-on-rtx-4000-ada-20gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: