Can Qwen 2.5 32B run on NVIDIA L4 24GB?

BARELY — Tight on Memory

B69Good

○Estimated from fit model

Qwen 2.5 32B needs ~27.0 GB VRAM. NVIDIA L4 24GB has 24.0 GB. With Q4_K_M quantization, expect ~6 tok/s.

Runtime: OllamaCapacity: OffloadBandwidth: LowStack: BasicBottleneck: Host offload

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) — 27.0 GB, 6.3 tok/s, Very compromised (needs ~2.2 GB host RAM)

27.0 GB required24.0 GB available

113% VRAM needed

3.0 GB over capacity — needs offload or smaller quantization

Fit status

Very compromised (needs ~2.2 GB host RAM)

Decode

6.3 tok/s

TTFT

30721 ms

Safe context

Memory

27.0 GB / 24.0 GB

Offload

10%

Memory breakdown

Weights19.5 GB

KV Cache3.9 GB

Runtime1.2 GB

Headroom2.4 GB

See how fast it feels

See how fast it feelsQwen 2.5 32B on NVIDIA L4 24GB
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: 6.3 tok/s decode · 30.7s TTFT (warm) · 16 tok/s prefill

What limits this setup

It fits through host-memory offload, and offload is the main reason performance drops.

CPU or host-memory offload is active

About 10% of the working set spills out of accelerator memory, which usually hurts latency and sustained decode throughput.

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

Remove offload with more accelerator memory

Prioritize a GPU or unified-memory tier that fits the whole model natively. Removing offload usually helps more than small compute gains.

Buy headroom, not only minimum fit

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

Increase host RAM if you keep offloading

This setup may need roughly {ram} GB of extra host RAM just for the offloaded portion, before OS and other tools.

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	A	Runs with offload (needs ~0.8 GB host RAM)	7.4 tok/s	14309 ms	4K
Coding	B	Very compromised	5.8 tok/s	33178 ms	4K
Agentic Coding	F	Too heavy	4.7 tok/s	59371 ms	4K
Reasoning	B	Very compromised (needs ~2.2 GB host RAM)	6.3 tok/s	36306 ms	4K
RAG	F	Too heavy	4.7 tok/s	74213 ms

Quantization options

How Qwen 2.5 32B (32B params) fits at each quantization level on NVIDIA L4 24GB (24.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	12.5 GB	Low	A84
Q3_K_S	3	15.7 GB	Low	A83
NVFP4Best for your GPU

Get started

Copy-paste commands to run Qwen 2.5 32B on your machine.

Run

ollama run qwen2.5

Upgrade options

Hardware that runs Qwen 2.5 32B well

👁 NVIDIA
RTX 5090 32GBBudget pick

32 GB VRAM (+8)1792 GB/s (+1492)

Removes host-memory offload, which is usually the single biggest latency and throughput win.66.4 tok/s decode

Removes host-memory offload, which is usually the single biggest latency and throughput win.

Raises estimated decode speed by about 954%.

~$1,999 MSRP

👁 NVIDIA
RTX PRO 4500 Blackwell 32GBBest value

32 GB VRAM (+8)896 GB/s (+596)

Removes host-memory offload, which is usually the single biggest latency and throughput win.41.6 tok/s decode

Removes host-memory offload, which is usually the single biggest latency and throughput win.

Raises estimated decode speed by about 560%.

~$2,499 MSRP

👁 NVIDIA
RTX 5000 Ada 32GBNVIDIA upgrade

32 GB VRAM (+8)576 GB/s (+276)

Removes host-memory offload, which is usually the single biggest latency and throughput win.25.5 tok/s decode

Removes host-memory offload, which is usually the single biggest latency and throughput win.

Raises estimated decode speed by about 305%.

~$4,000 MSRP

Frequently asked questions

See all results for NVIDIA L4 24GB See all hardware for Qwen 2.5 32B

URL: https://willitrunai.com/can-run/qwen-2.5-32b-on-l4-24gb

⇱

Can Qwen 2.5 32B run on NVIDIA L4 24GB?

BARELY — Tight on Memory

Choose the run profile you care about

Memory breakdown

See how fast it feels

What limits this setup

Best improvement path

Performance by workload

Quantization options

Get started

Hardware that runs Qwen 2.5 32B well

Frequently asked questions