1344 GB/s (+798)
C
Raises estimated decode speed by about 255%.116.4 tok/s decode
Raises estimated decode speed by about 255%.
Adds memory headroom for longer context windows and future model growth.
~$4,999 MSRP
Can StarCoder2 15B run on MacBook Pro M4 Max 48GB?
YES — Runs Great
C51Usable
○Estimated from fit model
StarCoder2 15B needs ~18.1 GB VRAM. MacBook Pro M4 Max 48GB has 34.6 GB. With Q5_K_M quantization, expect ~33 tok/s.
Runtime: llama.cppCapacity: RoomyBandwidth: MediumStack: StandardBottleneck: 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) — 18.1 GB, 32.8 tok/s, Runs well
18.1 GB required34.6 GB available
52% VRAM used
Fit status
Runs well
Decode
32.8 tok/s
TTFT
5908 ms
Safe context
16K
Memory
18.1 GB / 34.6 GB
Memory breakdown
Weights10.8 GB
KV Cache1.2 GB
Runtime0.9 GB
Headroom5.2 GB
See how fast it feels
See how fast it feelsStarCoder2 15B on MacBook Pro M4 Max 48GB
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: 32.8 tok/s decode · 5.9s TTFT (warm) · 82 tok/s prefill
What limits this setup
This setup is broadly balanced for this model.
Shared-memory contention still exists
The OS, browser, and inference runtime all compete for the same physical memory pool, so real-world headroom is less forgiving than raw capacity suggests.
Best improvement path
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | C | Runs well | 32.5 tok/s | 3251 ms | 16K |
| Coding | C | Runs well | 32.5 tok/s | 5959 ms | 16K |
| Agentic Coding | C | Runs well | 32.5 tok/s | 8668 ms | 16K |
| Reasoning | C | Runs well | 32.5 tok/s | 7043 ms | 16K |
| RAG | C | Runs well | 32.5 tok/s | 10835 ms | 16K |
Quantization options
How StarCoder2 15B (15B params) fits at each quantization level on MacBook Pro M4 Max 48GB (34.6 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 5.9 GB | Low | C45 |
Q3_K_S | 3 | 7.4 GB | Low | C45 |
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