Mac Studio M3 Ultra 256GBBudget pick
256 GB Unified (+128)
C
Adds memory headroom for longer context windows and future model growth.42 tok/s decode
Adds memory headroom for longer context windows and future model growth.
~$6,999 MSRP
Can StarCoder2 3B run on Gaudi 3 128GB?
YES — Runs Great
C41Usable
○Estimated from fit model
StarCoder2 3B needs ~16.0 GB VRAM. Gaudi 3 128GB has 128.0 GB. With Q4_K_M quantization, expect ~42 tok/s.
Runtime: llama.cppCapacity: RoomyBandwidth: HighStack: 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
Q4_K_M (Medium quality) — 16.0 GB, 42.0 tok/s, Runs well
16.0 GB required128.0 GB available
13% VRAM used
Fit status
Runs well
Decode
42.0 tok/s
TTFT
4610 ms
Safe context
16K
Memory
16.0 GB / 128.0 GB
Memory breakdown
Weights1.8 GB
KV Cache0.5 GB
Runtime0.9 GB
Headroom12.8 GB
See how fast it feels
See how fast it feelsStarCoder2 3B on Gaudi 3 128GB
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: 42.0 tok/s decode · 4.6s TTFT (warm) · 105 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 | Runs well | 42.0 tok/s | 2514 ms | 16K |
| Coding | C | Runs well | 42.0 tok/s | 4610 ms | 16K |
| Agentic Coding | C | Runs well | 42.0 tok/s | 6705 ms | 16K |
| Reasoning | C | Runs well | 42.0 tok/s | 5448 ms | 16K |
| RAG | C | Runs well | 42.0 tok/s | 8381 ms | 16K |
Quantization options
How StarCoder2 3B (3B params) fits at each quantization level on Gaudi 3 128GB (128.0 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.2 GB | Low | D37 |
Q3_K_S | 3 | 1.5 GB | Low | D37 |
NVFP4 | 4 | 1.7 GB | Medium | D37 |
Q4_K_M | 4 | 1.8 GB | Medium | D37 |
Q5_K_M | 5 | 2.2 GB | High | D37 |
Q6_K | 6 | 2.5 GB | High | D37 |
Q8_0 | 8 | 3.2 GB | Very High | D37 |
F16Best for your GPU | 16 | 6.1 GB | Maximum | D37 |
Get started
Copy-paste commands to run StarCoder2 3B on your machine.
Run
ollama run starcoder2:3bUpgrade options