24 GB VRAM (+8)456 GB/s (+232)
C
Raises estimated decode speed by about 103%.25.4 tok/s decode
Raises estimated decode speed by about 103%.
Adds memory headroom for longer context windows and future model growth.
~$599 MSRP
Can StarCoder2 15B run on Intel Arc Pro B50 16GB?
YES — Tight Fit
C49Usable
○Estimated from fit model
StarCoder2 15B needs ~14.5 GB VRAM. Intel Arc Pro B50 16GB has 16.0 GB. With Q5_K_M quantization, expect ~11 tok/s.
Runtime: llama.cppCapacity: TightBandwidth: Very lowStack: StandardBottleneck: Memory bandwidth
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) — 14.5 GB, 12.5 tok/s, Tight fit
14.5 GB required16.0 GB available
91% VRAM used
Fit status
Tight fit
Decode
12.5 tok/s
TTFT
15524 ms
Safe context
16K
Memory
14.5 GB / 16.0 GB
Memory breakdown
Weights10.8 GB
KV Cache1.2 GB
Runtime0.9 GB
Headroom1.6 GB
See how fast it feels
See how fast it feelsStarCoder2 15B on Intel Arc Pro B50 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: 12.5 tok/s decode · 15.5s TTFT (warm) · 31 tok/s prefill
What limits this setup
The raw memory story may look fine, but the software ecosystem is still a constraint here.
Runtime ecosystem is narrower than CUDA
Intel GPUs can look attractive on memory per dollar, but local AI tooling, kernels, and model coverage are still broader and easier on CUDA today.
Best improvement path
Prefer CUDA if you want the path of least resistance
If your goal is maximum runtime coverage, easier troubleshooting, and better support for new local AI releases, CUDA is usually still the safer upgrade path.
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | C | Tight fit | 12.5 tok/s | 8468 ms | 16K |
| Coding | C | Tight fit | 11.4 tok/s | 16947 ms | 16K |
| Agentic Coding | C | Runs with offload | 11.4 tok/s | 24650 ms | 16K |
| Reasoning | C | Tight fit | 12.5 tok/s | 18346 ms | 16K |
| RAG | C | Runs with offload | 11.4 tok/s | 30812 ms | 16K |
Quantization options
How StarCoder2 15B (15B params) fits at each quantization level on Intel Arc Pro B50 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