BBrainOutput
NVIDIA·PlatformPlatform target

NVIDIA GB10 (DGX Spark class): Specs & Local-AI Compatibility

Grace-Blackwell desktop AI platform with ~128GB unified memory — the DGX Spark class.

Some details here are provisional (Platform target). Treat specs as approximate and verify against the manufacturer before relying on them or purchasing.

Specs

Memory
128 GB unified
Memory type
Unified LPDDR5X
Bandwidth
273 GB/s
Approx FP16
to verify
Architecture
Grace Blackwell
Process
TSMC
Power
170 W
Launch
2025

Models this chip can run

Open models graded for a single NVIDIA GB10 (DGX Spark class), best fit first.

  • CodeLlama 13B
    CodeLlama · ~13B · 16K ctx · Llama Community License

    Fits at FP16 (~26GB) with ~63.6GB headroom — about 3 concurrent instances.

    FP16 · ~26GBRuns well
  • Gemma 3 12B
    Gemma 3 · ~12B · 128K ctx · Gemma Terms of Use

    Fits at FP16 (~24GB) with ~65.6GB headroom — about 3 concurrent instances.

    FP16 · ~24GBRuns well
  • Mistral Nemo 12B
    Mistral · ~12B · 128K ctx · Apache-2.0

    Fits at FP16 (~24GB) with ~65.6GB headroom — about 3 concurrent instances.

    FP16 · ~24GBRuns well
  • Gemma 2 9B
    Gemma · ~9B · 8K ctx · Gemma Terms of Use

    Fits at FP16 (~19GB) with ~70.6GB headroom — about 4 concurrent instances.

    FP16 · ~19GBRuns well
  • Llama 3.1 8B
    Llama · ~8B · 128K ctx · Llama Community License

    Fits at FP16 (~17GB) with ~72.6GB headroom — about 5 concurrent instances.

    FP16 · ~17GBRuns well
  • Qwen3 8B
    Qwen · ~8B · 128K ctx · Apache-2.0

    Fits at FP16 (~17GB) with ~72.6GB headroom — about 5 concurrent instances.

    FP16 · ~17GBRuns well
  • Granite 3 8B
    Granite · ~8B · 128K ctx · Apache-2.0

    Fits at FP16 (~17GB) with ~72.6GB headroom — about 5 concurrent instances.

    FP16 · ~17GBRuns well
  • DeepSeek-R1 Distill 8B
    DeepSeek · ~8B · 128K ctx · MIT

    Fits at FP16 (~17GB) with ~72.6GB headroom — about 5 concurrent instances.

    FP16 · ~17GBRuns well

Devices built on this chip

Build a private AI Business OS on NVIDIA GB10 (DGX Spark class)

Run your own AI agents on hardware you control — private by design, no per-seat data leaving your premises. BrainOutput helps you pick the right machine and turn it into a working AI Business OS.

Explore the AI Business OS