BBrainOutput
NVIDIA·PlatformAnnounced

NVIDIA GB200 (Grace-Blackwell): Specs & Local-AI Compatibility

Grace-Blackwell superchip platform for rack-scale AI. Placeholder.

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

Specs

Memory
384 GB unified
Memory type
HBM3e + LPDDR5X
Bandwidth
to verify
Approx FP16
to verify
Architecture
Grace Blackwell
Process
TSMC 4NP
Power
to verify
Launch
2025

Models this chip can run

Open models graded for a single NVIDIA GB200 (Grace-Blackwell), best fit first.

  • Gemma 2 27B
    Gemma · ~27B · 8K ctx · Gemma Terms of Use

    Fits at FP16 (~54GB) with ~214.8GB headroom — about 4 concurrent instances.

    FP16 · ~54GBRuns well
  • Gemma 3 27B
    Gemma 3 · ~27B · 128K ctx · Gemma Terms of Use

    Fits at FP16 (~54GB) with ~214.8GB headroom — about 4 concurrent instances.

    FP16 · ~54GBRuns well
  • Mistral Small 24B
    Mistral · ~24B · 32K ctx · Apache-2.0

    Fits at FP16 (~48GB) with ~220.8GB headroom — about 5 concurrent instances.

    FP16 · ~48GBRuns well
  • DeepSeek-Coder V2 (class)
    DeepSeek · ~16B · 128K ctx · DeepSeek License

    Fits at FP16 (~33GB) with ~235.8GB headroom — about 8 concurrent instances.

    FP16 · ~33GBRuns well
  • StarCoder2 15B
    StarCoder · ~15B · 16K ctx · BigCode OpenRAIL-M

    Fits at FP16 (~30GB) with ~238.8GB headroom — about 8 concurrent instances.

    FP16 · ~30GBRuns well
  • Qwen2.5 14B
    Qwen · ~14B · 128K ctx · Apache-2.0

    Fits at FP16 (~30GB) with ~238.8GB headroom — about 8 concurrent instances.

    FP16 · ~30GBRuns well
  • Qwen3 14B
    Qwen · ~14B · 128K ctx · Apache-2.0

    Fits at FP16 (~30GB) with ~238.8GB headroom — about 8 concurrent instances.

    FP16 · ~30GBRuns well
  • Phi-3 Medium (14B)
    Phi · ~14B · 128K ctx · MIT

    Fits at FP16 (~28GB) with ~240.8GB headroom — about 9 concurrent instances.

    FP16 · ~28GBRuns well

Build a private AI Business OS on NVIDIA GB200 (Grace-Blackwell)

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