Mac Studio M3 Ultra 256GBコスパ重視
256 GB Unified (+128)819 GB/s (+546)
B
Adds memory headroom for longer context windows and future model growth.4.7 tok/sデコード
Adds memory headroom for longer context windows and future model growth.
〜$6,999 MSRP
Can BGE Large EN v1.5 run on NVIDIA DGX Spark 128GB?
YES — With F16
B67Good
Estimated from fit model
BGE Large EN v1.5 needs ~16.4 GB VRAM. NVIDIA DGX Spark 128GB has 0 MB. With F16 quantization, expect ~5 tok/s.
Runtime: OllamaCapacity: RoomyBandwidth: LowStack: BasicBottleneck: Memory bandwidth
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.
BGE Large EN v1.5 at F16 needs 2.9 GB — too much for NVIDIA DGX Spark 128GB (0.0 GB). Runs at F16 (16.4 GB) with maximum quality. 8 quantization levels fit.
Select quantization to explore
F16 (Maximum quality) — 16.4 GB, 4.7 tok/s, Runs well
16.4 GB required108.8 GB available
15% VRAM used
Fit status
Runs well
Decode
4.7 tok/s
TTFT
41279 ms
Safe context
512
Memory
16.4 GB / 108.8 GB
Memory breakdown
Weights0.7 GB
KV Cache1.5 GB
Runtime1.2 GB
Headroom13.1 GB
See how fast it feels
See how fast it feelsBGE Large EN v1.5 on NVIDIA DGX Spark 128GB
1st promptCold start — includes initialization
>What is local AI inference?
Local AI inference means running an AI language model directly on your own hardware — your laptop, desktop, or server — instead of sending requests to a remote cloud API.
When you run inference locally the model weights are loaded into your GPU or unified memory. Each token you generate requires reading those weights from memory, so memory bandwidth is the main bottleneck for decode speed.
Key benefits of running locally:
- Full privacy: your prompts never leave your machine
- No per-token cost or rate limits
- Works offline once the model is downloaded
- Latency depends only on your hardware
2nd promptWarm — model ready, no init overhead
>How much VRAM do I need?
It depends on the model size and quantization level. A rough rule of thumb:
Model size Q4 (4-bit) Q8 (8-bit) FP16
7B params ~4.3 GB ~7.5 GB ~14 GB
13B params ~7.9 GB ~13.9 GB ~26 GB
70B params ~42.7 GB ~74.9 GB ~140 GB
Most people use 4-bit quantization (Q4_K_M) which gives 90-95% of full quality at a fraction of the memory. A 24 GB GPU can comfortably run most 7B-13B models.
Estimated: 4.7 tok/s decode · 41.3s TTFT (warm) · 12 tok/s prefill
What limits this setup
The model fits in shared memory, but shared-memory bandwidth is now the real limiter.
Fit does not mean dedicated-VRAM speed
Unified or shared memory can make a model technically fit, but sustained tokens per second may still trail a discrete high-bandwidth GPU with less total memory.
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
Prioritize bandwidth, not only capacity
If this workload feels slow, the next useful step is often a GPU tier with materially faster memory bandwidth rather than only a small bump in capacity.
Performance by workload
| Workload | Grade | Fit | Decode | TTFT | Context |
|---|---|---|---|---|---|
| Chat | B | Runs well | 4.7 tok/s | 22516 ms | 512 |
| Coding | F | Too heavy | 4.7 tok/s | 41279 ms | 512 |
| Agentic Coding | B | Runs well | 4.7 tok/s | 60043 ms | 512 |
| Reasoning | B | Runs well | 4.7 tok/s | 48785 ms | 512 |
| RAG | B | Runs well | 4.7 tok/s | 75053 ms | 512 |
Quantization options
How BGE Large EN v1.5 (0.33500000834465027B params) fits at each quantization level on NVIDIA DGX Spark 128GB (92.2 GB usable).
| Quant | Bits | VRAM | Quality | Fit |
|---|---|---|---|---|
Q2_K | 2 | 0.1 GB | Low | A71 |
Q3_K_S | 3 | 0.2 GB | Low | A71 |
NVFP4 | 4 | 0.2 GB | Medium | A71 |
Q4_K_M | 4 | 0.2 GB | Medium | A71 |
Q5_K_M | 5 | 0.2 GB | High | A71 |
Q6_K | 6 | 0.3 GB | High | A71 |
Q8_0 | 8 | 0.4 GB | Very High | A71 |
F16Best for your GPU | 16 | 0.7 GB | Maximum | A71 |
Get started
Copy-paste commands to run BGE Large EN v1.5 on your machine.
Run
docker run --rm -it ghcr.io/ggerganov/llama.cpp:full \
--hf-repo "BAAI/bge-large-en-v1.5" \
--hf-file "bge-large-en-v1.5-F16.gguf" \
-c 4096 -ngl 99アップグレードオプション
BGE Large EN v1.5を快適に動かすハードウェア
Frequently asked questions
Can NVIDIA DGX Spark 128GB run BGE Large EN v1.5?
Yes, NVIDIA DGX Spark 128GB can run BGE Large EN v1.5 at F16 quantization (Runs well). The recommended F16 requires 2.9 GB which exceeds available memory, but at F16 it needs only 16.4 GB. Expected decode speed: 4.7 tok/s.
How much VRAM does BGE Large EN v1.5 need?
BGE Large EN v1.5 (0.33500000834465027B parameters) requires approximately 2.9 GB at F16 quantization. On NVIDIA DGX Spark 128GB, it fits at F16 using 16.4 GB.
What is the best quantization for BGE Large EN v1.5?
The recommended quantization is F16, but on NVIDIA DGX Spark 128GB the best fitting quantization is F16, which uses 16.4 GB.
What speed will BGE Large EN v1.5 run at on NVIDIA DGX Spark 128GB?
On NVIDIA DGX Spark 128GB, BGE Large EN v1.5 achieves approximately 4.7 tokens per second decode speed with a time-to-first-token of 41279ms using F16 quantization.
Can NVIDIA DGX Spark 128GB run BGE Large EN v1.5 for coding?
For coding workloads, BGE Large EN v1.5 on NVIDIA DGX Spark 128GB receives a F grade with 4.7 tok/s and 512 context.
What context window can BGE Large EN v1.5 use on NVIDIA DGX Spark 128GB?
On NVIDIA DGX Spark 128GB, BGE Large EN v1.5 can safely use up to 512 tokens of context at F16 quantization. The model's official context limit is 512, but available memory constrains the safe maximum.
What should I upgrade first if BGE Large EN v1.5 feels slow on NVIDIA DGX Spark 128GB?
Prioritize bandwidth, not only capacity. If this workload feels slow, the next useful step is often a GPU tier with materially faster memory bandwidth rather than only a small bump in capacity.
Is unified memory on NVIDIA DGX Spark 128GB as fast as VRAM for BGE Large EN v1.5?
Not always. NVIDIA DGX Spark 128GB can often fit larger models thanks to unified memory, but a discrete GPU with dedicated high-bandwidth VRAM may still decode faster once the model fits. For this combination, the important distinction is capacity versus sustained throughput.
Embed this result▼
Paste this snippet into any page to show a live fit card.
<iframe src="https://willitrunai.com/embed/bge-large-en-v1.5-on-dgx-spark-128gb" width="400" height="180" frameborder="0" style="border:none;border-radius:12px;overflow:hidden;" title="Will It Run AI — fit result"></iframe>
Preview: