Mac Studio M3 Ultra 256GBBudget pick
256 GB Unified (+128)819 GB/s (+546)
B
Adds memory headroom for longer context windows and future model growth.56 tok/s decode
Adds memory headroom for longer context windows and future model growth.
~$6,999 MSRP
Can Gemma 3 4B run on NVIDIA DGX Spark 128GB?
YES — With F16
B64Good
○Estimated from fit model
Gemma 3 4B needs ~24.5 GB VRAM. NVIDIA DGX Spark 128GB has 0 MB. With F16 quantization, expect ~29 tok/s.
Runtime: OllamaCapacity: RoomyBandwidth: 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.
Gemma 3 4B at Q4_K_M needs 5.7 GB — too much for NVIDIA DGX Spark 128GB (0.0 GB). Runs at F16 (24.5 GB) with maximum quality. 8 quantization levels fit.
Select quantization to explore
Q4_K_M (Medium quality) — 18.8 GB, 56.0 tok/s, Runs well
18.8 GB required108.8 GB available
17% VRAM used
Fit status
Runs well
Decode
56.0 tok/s
TTFT
3457 ms
Safe context
128K
Memory
18.8 GB / 108.8 GB
Memory breakdown
Weights2.4 GB
KV Cache2.1 GB
Runtime1.2 GB
Headroom13.1 GB
See how fast it feels
See how fast it feelsGemma 3 4B on NVIDIA DGX Spark 128GB
1st promptCold start — includes initialization
>Solve: If a train travels at 60 mph for 2.5 hours, how far does it go?
Let me work through this step by step.
Given information:
- Speed = 60 mph (miles per hour)
- Time = 2.5 hours
Formula: Distance = Speed x Time
Calculation:
Distance = 60 mph x 2.5 h
= 150 miles
The train travels 150 miles.
Double-check: at 60 mph the train covers 60 miles each hour. In half an hour it covers 30 miles. So in 2.5 hours: 60 + 60 + 30 = 150 miles. Confirmed.
2nd promptWarm — model ready, no init overhead
>What if the train stops for 20 minutes halfway?
The stop doesn't change the total distance covered while moving, but it affects total trip time.
Breakdown:
- First half: 1.25 hours at 60 mph = 75 miles
- Stop: 20 minutes (no distance)
- Second half: 1.25 hours at 60 mph = 75 miles
Total distance: 150 miles (same)
Total time: 2.5 h + 0.33 h = 2.83 hours
Average speed: 150 / 2.83 = 53 mph
The distance stays the same but average speed drops to 53 mph because of the stop.
Estimated: 56.0 tok/s decode · 3.5s TTFT (warm) · 140 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 | F | Too heavy | 12.1 tok/s | 8739 ms | 4K |
| Coding | F | Too heavy | 12.1 tok/s | 16022 ms | 4K |
| Agentic Coding | F | Too heavy | 12.1 tok/s | 23304 ms | 4K |
| Reasoning | F | Too heavy | 12.1 tok/s | 18935 ms | 4K |
| RAG | F | Too heavy | 12.1 tok/s | 29130 ms | 4K |
Quantization options
How Gemma 3 4B (4B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 1.6 GB | Low | B60 |
Q3_K_S | 3 | 2.0 GB | Low | B60 |
NVFP4 | 4 |
Get started
Copy-paste commands to run Gemma 3 4B on your machine.
Run
ollama run gemma3:4bUpgrade options