Can StarCoder2 15B run on Tesla P40 24GB?

YES — Runs Great

C52Usable

Estimated from fit model

StarCoder2 15B needs ~15.6 GB VRAM. Tesla P40 24GB has 24.0 GB. With Q5_K_M quantization, expect ~21 tok/s.

Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: 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

Q5_K_M (High quality) — 15.6 GB, 21.0 tok/s, Runs well

15.6 GB required24.0 GB available

65% VRAM used

Fit status

Runs well

Decode

21.0 tok/s

TTFT

9198 ms

Safe context

16K

Memory

15.6 GB / 24.0 GB

Memory breakdown

Weights10.8 GB

KV Cache1.2 GB

Runtime1.2 GB

Headroom2.4 GB

See how fast it feels

See how fast it feelsStarCoder2 15B on Tesla P40 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: 21.0 tok/s decode · 9.2s TTFT (warm) · 53 tok/s prefill

What limits this setup

This setup is broadly balanced for this model.

Older PCIe generation

PCIe 3.0 is workable, but it compounds the penalty when you offload heavily or try to scale across multiple cards.

Best improvement path

Performance by workload

Workload	Grade	Fit	Decode	TTFT	Context
Chat	C	Runs well	21.0 tok/s	5017 ms	16K
Coding	C	Runs well	21.0 tok/s	9198 ms	16K
Agentic Coding	C	Runs well	21.0 tok/s	13379 ms	16K
Reasoning	C	Runs well	21.0 tok/s	10871 ms	16K
RAG	C	Runs well	21.0 tok/s	16724 ms	16K

Inference speed

StarCoder2 15B inference speed — tokens per second by GPU & Mac

Estimated decode speed (tokens/sec) for StarCoder2 15B at Q5_K_M across popular GPUs and Apple Silicon, using the fastest local runtime per device. Fastest is RTX 5090 32GB at ~124 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	Q5_K_M	123.8	Fits
RTX 4090 24GB	24 GB	Q5_K_M	79.0	Fits
RX 7900 XTX 24GB	24 GB	Q5_K_M	71.3	Fits
RTX 3090 24GB	24 GB	Q5_K_M	67.6	Fits
RTX 4080 Super 16GB	16 GB	Q5_K_M	64.4	Tight
Mac Studio M3 Ultra 256GB	256 GB	Q5_K_M	57.4	Fits
Mac Studio M2 Ultra 128GB	128 GB	Q5_K_M	47.8	Fits
Mac Studio M1 Ultra 128GB	128 GB	Q5_K_M	45.4	Fits
MacBook Pro M4 Max 128GB	128 GB	Q5_K_M	32.8	Fits
MacBook Pro M4 Max 64GB	64 GB	Q5_K_M	32.8	Fits
MacBook Pro M3 Max 64GB	64 GB	Q5_K_M	24.7	Fits
MacBook Pro M1 Max 64GB	64 GB	Q5_K_M	22.7	Fits
RTX 4070 12GB	12 GB	Q5_K_M	21.2	Heavy offload
MacBook Pro M4 Pro 48GB	48 GB	Q5_K_M	20.0	Fits
RTX 3060 12GB	12 GB	Q5_K_M	12.5	Heavy offload
RTX 4060 8GB	8 GB	Q5_K_M	4.7	Too big

Estimates for single-stream decoding at Q5_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 StarCoder2 15B (15B params) fits at each quantization level on Tesla P40 24GB (24.0 GB usable).

Quant	Bits	VRAM	Quality	Fit
Q2_K	2	5.9 GB	Low	C47
Q3_K_S	3	7.4 GB	Low	C48
NVFP4	4	8.4 GB	Medium	C49
Q4_K_M	4	9.2 GB	Medium	C49
Q5_K_M	5	10.8 GB	High	C51
Q6_K	6	12.3 GB	High	C52
Q8_0Best for your GPU	8	16.1 GB	Very High	C51
F16	16	30.7 GB	Maximum	F0

Get started

Copy-paste commands to run StarCoder2 15B on your machine.

Run

docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
  --hf-repo "bigcode/starcoder2-15b" \
  --hf-file "starcoder2-15b-Q5_K_M.gguf" \
  -c 4096 -ngl 99

Opciones de mejora

Hardware que ejecuta bien StarCoder2 15B

RTX 5090 32GBOpción económica

32 GB VRAM (+8)1792 GB/s (+1446)

Sube la velocidad estimada de decodificación alrededor de un 490%.123.8 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 490%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$1,999 MSRP

RTX PRO 4500 Blackwell 32GBMejor relación calidad-precio

32 GB VRAM (+8)896 GB/s (+550)

Sube la velocidad estimada de decodificación alrededor de un 270%.77.6 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 270%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$2,499 MSRP

NVIDIA V100 32GBMejora NVIDIA

32 GB VRAM (+8)900 GB/s (+554)

Sube la velocidad estimada de decodificación alrededor de un 196%.62.2 tok/s decodificación

Sube la velocidad estimada de decodificación alrededor de un 196%.

Añade margen de memoria para más contexto y para que el modelo envejezca mejor.

~$8,999 MSRP

Frequently asked questions

Can Tesla P40 24GB run StarCoder2 15B?

Yes, Tesla P40 24GB can run StarCoder2 15B with a C grade (Runs well). Expected decode speed: 21.0 tok/s.

How much VRAM does StarCoder2 15B need?

StarCoder2 15B (15B parameters) requires approximately 15.6 GB of memory with Q5_K_M quantization.

What is the best quantization for StarCoder2 15B?

The recommended quantization for StarCoder2 15B is Q5_K_M, which balances quality and memory efficiency.

What speed will StarCoder2 15B run at on Tesla P40 24GB?

On Tesla P40 24GB, StarCoder2 15B achieves approximately 21.0 tokens per second decode speed with a time-to-first-token of 9198ms using Q5_K_M quantization.

Can Tesla P40 24GB run StarCoder2 15B for coding?

For coding workloads, StarCoder2 15B on Tesla P40 24GB receives a C grade with 21.0 tok/s and 16K context.

What context window can StarCoder2 15B use on Tesla P40 24GB?

On Tesla P40 24GB, StarCoder2 15B can safely use up to 16K tokens of context. The model's official context limit is 16K, but available memory constrains the safe maximum.

See all results for Tesla P40 24GB See all hardware for StarCoder2 15B

Embed this result▼

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

<iframe src="https://willitrunai.com/embed/starcoder2-15b-on-tesla-p40-24gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>

Preview: