DEPLOYDatabase

AI chip · NVIDIA

NVIDIA GB200 Grace Blackwell Superchip

Two B200 GPUs + one Grace CPU on a single board via NVLink-C2C. Base unit of the GB200 NVL72 rack. The flagship 2024-2025 AI-training package.

On the DEPLOY graph: 11 campuses carry a cited inventory line for it · 1 reference-design cluster include it.

Market position

GB200 is the Blackwell Superchip: 2 B200 GPUs coherently packaged with 1 Grace CPU over 900 GB/s NVLink-C2C. Sold as the NVL72 rack (72 GPUs + 36 Graces). This is the current top-of-stack training chip: Stargate, Colossus 2, Meta Hyperion, HUMAIN all committed to it.

DEPLOY editorial. What the vendor PDF cannot tell you.

Physical-AI cross-link

Derived from claims already on record (vendor first-party TDP, throughput, price, and the reference-design cluster registry). The arithmetic is shown per row so it can be audited. No estimated inputs.

Perf per watt
3.33 FP8 TFLOPS/W
9,000 TFLOPS ÷ 2700 W = 3.33
Rack power (in this design)
132 kW per NVIDIA GB200 NVL72 (~3667 W per chip incl. overhead)
132 kW ÷ 36 chips = 3667 W each (system-level, includes CPU/mem/NIC/cooling)
Cooling class (across reference designs)
water
Observed across 1 reference-design cluster containing this chip

What fits in 384 GB

Weights-only footprint for public open-weight LLMs. Pure arithmetic: params × bytes/param. Excludes KV cache and activation memory; add ~10-30% headroom for real serving. A model that does not fit at FP16 may still fit at INT8 or INT4 with quality trade-offs. Not a benchmark.

ModelParamsFP16INT8INT4
Llama 3.1 8B
dense
8 B16 GB ✓8 GB ✓4 GB ✓
Llama 3.1 70B
dense
70 B140 GB ✓70 GB ✓35 GB ✓
Llama 3.1 405B
dense
405 B810 GB ✗405 GB ✗203 GB ✓
Llama 3.3 70B
dense
70 B140 GB ✓70 GB ✓35 GB ✓
DeepSeek V3
MoE (671B total, 37B active per token)
671 B1342 GB ✗671 GB ✗336 GB ✓
DeepSeek R1
MoE (671B total, 37B active per token)
671 B1342 GB ✗671 GB ✗336 GB ✓
Qwen 2.5 7B
dense
7 B14 GB ✓7 GB ✓4 GB ✓
Qwen 2.5 72B
dense
72 B144 GB ✓72 GB ✓36 GB ✓
Mixtral 8x7B
MoE (46.7B total, 12.9B active per token)
46.7 B93 GB ✓47 GB ✓23 GB ✓
Mixtral 8x22B
MoE (141B total, 39B active per token)
141 B282 GB ✓141 GB ✓71 GB ✓
Gemma 2 27B
dense
27 B54 GB ✓27 GB ✓14 GB ✓
Command R+
dense
104 B208 GB ✓104 GB ✓52 GB ✓
Kimi K2
MoE (1T total, 32B active per token)
1000 B2000 GB ✗1000 GB ✗500 GB ✗

Math: FP16 = params × 2 bytes; INT8 = params × 1 byte; INT4 = params × 0.5 bytes. MoE models sum every expert (full weights on disk), not the per-token active subset.

Common questions

Answer-first, sourced. Every claim below traces to a specific field on this page or a cited datasheet. FAQPage schema is emitted so LLM crawlers can lift these verbatim.

How much does NVIDIA GB200 Grace Blackwell Superchip cost?

NVIDIA GB200 Grace Blackwell Superchip launch pricing is not on record in the DEPLOY registry. Vendor list prices are typically published in the vendor's own briefing rather than the datasheet.

How much memory does NVIDIA GB200 Grace Blackwell Superchip have?

384 GB of HBM3e at 16,000 GB/s. Vendor datasheet, first-party.

How much power does one NVIDIA GB200 Grace Blackwell Superchip draw?

2700 W TDP (thermal design power) per chip. System-level draw is higher: see the reference-design section for rack-level kW.

Which data centers have NVIDIA GB200 Grace Blackwell Superchip?

11 campuses on the DEPLOY registry carry a cited inventory line for this chip. Full list below with per-campus quantity hints where published.

Which open-weight LLMs fit on one NVIDIA GB200 Grace Blackwell Superchip?

Of the 13 public open-weight LLMs DEPLOY tracks: 9 fit at FP16, 9 at INT8, 12 at INT4 (weights only, excludes KV cache). The full table is above with per-model math. Sparsity, offloading and multi-GPU serving change the picture; this row is single-chip weights-only.

Key facts

Class
Compute SoC (AI accelerator)
Designer
NVIDIA
Safety-critical?
No (data-center inference / training)
Record as of
2026-08-25
Latest cited claim
2025-07-14 (across 12 inventory + benchmark rows)
Specifications (8 fields, click to expand)

Vendor datasheet figures (first-party). Dense throughput first; sparse (2:4) in parentheses where NVIDIA quotes it. The exhaustive spec sheet lives on the datasheet URL below the table: this row set covers what buyers actually compare on.

Process node
TSMC 4NP
TDP
2700 W
Memory
384 GB HBM3e
Memory bandwidth
16,000 GB/s
FP8 (dense)
9,000 TFLOPS (18,000 sparse)
Form factor
Superchip (2xB200 + 1xGrace)
Announced
2024-03-18
Released
2024-12-01

Source: vendor datasheet

Generation

Used in reference designs

The named rack-scale and pod-scale designs buyers actually order that contain this chip.

See all reference designs →

Compare with

See every chip comparison →

Benchmarks

Published measurements per workload. Vendor datasheet numbers are first-party (verified posture); MLPerf / InferenceX / press results come in when their license terms + workload naming permit.

WorkloadValueUnitSourceAs of
peak tflops fp4 dense(Two B200 GPUs per superchip)10000TFLOPSvendor2024-03-01

Data centers stocking NVIDIA GB200 Grace Blackwell Superchip

Who has the most? →

Every campus on the DEPLOY registry with a public claim of NVIDIA GB200 Grace Blackwell Superchip on site. Quantity hints are unstructured because underlying disclosures vary by order of magnitude.

  1. partially energizedreported

    NVIDIA Blackwell training capacity at the 1 GW Prometheus site

    Meta

  2. Cologix Johnstown Campusas of 2025-06-01
    announcedinferred

    NVIDIA Blackwell capacity for AI-training tenants at the Johnstown campus

    Multi-tenant hyperscale campus; NVIDIA Blackwell tenants inferred from CoreWeave-generation posture.

    Cologix

  3. under constructionreported

    Planned Blackwell capacity for the Fermi HyperGrid campus (2 GW target, four planned SMRs)

    Fermi America

  4. under constructionreported

    Up to 500,000 Blackwell GPUs committed across the HUMAIN Riyadh/Dammam campuses (multi-year)

    NVIDIA

  5. Homer City Energy Campusas of 2025-04-01
    under constructioninferred

    Planned 4.5 GW gas-fed AI campus targeting NVIDIA Blackwell tenants

    Announced as an AI-first campus. Chip mix not yet published; Blackwell inferred from AI-training positioning.

    Homer City Redevelopment

  6. Stargate Abileneas of 2025-01-21
    partially energizedreported

    GB200 NVL72 racks part of the Blackwell buildout

    OpenAI

  7. energizedinferred

    Blackwell-generation training capacity at Cheyenne

    Site published as Fairwater-class; specific chip inferred from Microsoft's Blackwell-first stance for new build

    Microsoft

  8. partially energizedreported

    Second Fairwater campus targeting Blackwell-generation compute

    Microsoft

  9. partially energizedreported

    Blackwell-generation AI datacenter (Fairwater flagship); Microsoft's next-generation closed-loop cooled AI DC design

    Microsoft

  10. announcedinferred

    Microsoft-anchored AI capacity at the Chevron gas-turbine-fed campus

    Inferred from Microsoft's Fairwater-generation Blackwell posture; site is a Chevron-Microsoft build for AI training

    Microsoft

  11. announcedreported

    1.5 GW zero-carbon AI factory targeting NVIDIA Blackwell tenants

    DataVolt

Sources

Adoption rows appear as we verify chip-integration claims per model. Every row is a public claim the maker or a tier-1 source has stated; verification tier (verified / reported / inferred) is shown per row. See every chip on record for the full catalog or the NVIDIA record.