Question 1

What AI models can run on DGX Spark (GB10)?

Accepted Answer

The DGX Spark (GB10) (128 GB VRAM) can run models including Llama-2-7b-chat, Llama-3.2-1B-Instruct-QLORA_INT4_EO8, MediPhi-Instruct, Llama-3.2-3B-Instruct-SpinQuant_INT4_EO8, Llama-3.2-1B-Instruct-SpinQuant_INT4_EO8 and more — 10 compatible models indexed so far. Compatible models are determined by whether the model's memory footprint fits within the GPU's 128 GB VRAM, potentially using quantization (Q8, Q6, or Q4) to reduce memory usage.

Question 2

What is the cheapest cloud option for DGX Spark (GB10)?

Accepted Answer

We don't currently have cloud pricing data for the DGX Spark (GB10). This GPU may not be widely available from major cloud providers, or pricing data hasn't been synced yet.

Question 3

Is DGX Spark (GB10) good for fine-tuning?

Accepted Answer

Yes, the DGX Spark (GB10) is excellent for fine-tuning. With 128 GB of VRAM, it can handle full fine-tuning of models up to ~30B parameters and LoRA/QLoRA fine-tuning of much larger models. It's a top-tier choice for training workloads.

Question 4

How much VRAM does DGX Spark (GB10) have?

Accepted Answer

The DGX Spark (GB10) has 128 GB of VRAM (LPDDR5X). It also offers 273 GB/s memory bandwidth, 125 TFLOPS of FP16 compute, FP8 support for efficient inference.

Question 5

What quantization formats work on DGX Spark (GB10)?

Accepted Answer

The DGX Spark (GB10) supports the following quantization formats: FP16 (full precision), FP8, Q8 (8-bit), Q6 (6-bit), Q4 (4-bit). FP16 works for models up to ~40B parameters. FP8 halves memory vs FP16 with minimal quality loss. Q4 quantization lets models up to ~435B parameters (by VRAM estimate) fit on the DGX Spark (GB10).

DGX Spark (GB10)

Specifications

Cloud pricing

Compatible models

Compare this GPU

Frequently asked questions

VRAM	128 GB
Memory type	LPDDR5X
Bandwidth	273 GB/s
FP16 TFLOPS	125
FP8 support	Yes