20 GB VRAM (+4)
C
Adds memory headroom for longer context windows and future model growth.29 tok/s decode
Adds memory headroom for longer context windows and future model growth.
~$1,250 MSRP
Can StarCoder2 15B run on RTX A4000 16GB?
YES — Tight Fit
C52Usable
○Estimated from fit model
StarCoder2 15B needs ~14.8 GB VRAM. RTX A4000 16GB has 16.0 GB. With Q5_K_M quantization, expect ~30 tok/s.
Runtime: OllamaCapacity: TightBandwidth: LowStack: BasicBottleneck: Balanced
Share:
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.
Select quantization to explore
Q5_K_M (High quality) — 14.8 GB, 32.3 tok/s, Tight fit
14.8 GB required16.0 GB available
93% VRAM used
Fit status
Tight fit
Decode
32.3 tok/s
TTFT
5988 ms
Safe context
16K
Memory
14.8 GB / 16.0 GB
Memory breakdown
Weights10.8 GB
KV Cache1.2 GB
Runtime1.2 GB
Headroom1.6 GB
See how fast it feels
See how fast it feelsStarCoder2 15B on RTX A4000 16GB
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: 32.3 tok/s decode · 6.0s TTFT (warm) · 81 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 | C | Tight fit | 29.6 tok/s | 3565 ms | 16K |
| Coding | C | Tight fit | 29.6 tok/s | 6537 ms | 16K |
| Agentic Coding | C | Runs with offload | 22.1 tok/s | 12746 ms | 16K |
| Reasoning | C | Tight fit | 29.6 tok/s | 7725 ms | 16K |
| RAG | C | Runs with offload | 22.1 tok/s | 15933 ms | 16K |
Quantization options
How StarCoder2 15B (15B params) fits at each quantization level on RTX A4000 16GB (16.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.9 GB | Low | C51 |
Q3_K_S | 3 | 7.4 GB | Low | C53 |
NVFP4 | 4 |
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 99Upgrade options
Hardware that runs StarCoder2 15B well
24 GB VRAM (+8)936 GB/s (+488)
B
Raises estimated decode speed by about 109%.67.6 tok/s decode
Raises estimated decode speed by about 109%.
Adds memory headroom for longer context windows and future model growth.
~$1,499 MSRP
24 GB VRAM (+8)1008 GB/s (+560)
B
Raises estimated decode speed by about 145%.79 tok/s decode
Raises estimated decode speed by about 145%.
Adds memory headroom for longer context windows and future model growth.
~$1,599 MSRP