SECTIONS
EDITIONS
NETWORK
AIXX Pedia The encyclopedia of compute power

AIXX Pedia › Model Entry › Intel Gaudi 3

Data center accelerator illustrative image (AI-generated · not an official product photo)
Illustrative image (AI-generated, not an official product image) · AIXX Pedia

Intel Gaudi 3

FP8 1835 TFLOPS · 128GB HBM2e · 24×200GbE · 900W OAM · Model Entry

Overview OVERVIEW

Intel Gaudi 3 Intel's third-generation data center AI accelerator line (released April 2024, shipped in the second half of the same year · per official statements): FP8 1835 TFLOPS 、 128GB HBM2e · 3.7TB/s , on-card 24× 200GbE Ethernet RDMA, OAM form factor 900W (official statements). Its differentiating core is Ethernet-based scale-out interconnect —using standard Ethernet stacks for cluster expansion, a distinct path from NVLink's proprietary interconnect (architectural perspective). Positioned against NVIDIA H100 (see comparison table for Intel's official figures).

History HISTORY

  • 2022-2023 : Gaudi 2 shipped at scale (Habana Labs line · official statements).
  • 2024-04 : Gaudi 3 launched (Intel Vision · official statements); OAM/PCIe dual form factors shipped in the second half of the year.
  • 2024-2025 : OEM integration and shipment by Dell/Supermicro and others (channel perspective).
  • 2025-01 : Intel canceled the Falcon Shores discrete product plan (public reporting)—Gaudi 3 became the final third-generation product of the line; software stack evolution follows Intel's current roadmap.

Gaudi Generational Comparison GENERATIONS

DimensionGaudi 2 (previous generation · official statements)Gaudi 3 (this entry · official statements)
FP8 Compute432 TFLOPS(MHE)1835 TFLOPS(MHE)
BF16216 TFLOPS735 TFLOPS
Memory96GB HBM2e · 2.45TB/s128GB HBM2e · 3.7TB/s
Interconnect21× 100GbE(RoCE)24× 200GbE(RoCE/RDMA)
Power (OAM)600W-class900W

Scope: all figures are Intel's official release statements (verified 2026-11). MHE = Matrix Heavy Engine compute basis; power figures are for the OAM form factor and differ for PCIe.

Technical Specifications SPECS

ItemSpecifications (Intel official statements)
FP81835 TFLOPS (MHE · official statements)
BF16735 TFLOPS (official statements)
Memory128GB HBM2e · 3.7TB/s (official statements)
Interconnect24× 200GbE integrated NICs · RoCE/RDMA scale-out (official statements)
Form factorOAM (900W) / PCIe dual form factors (official statements)
Software stackSynapseAI · oneAPI · PyTorch/Vision support (official statements)
OEM IntegrationDell / Supermicro / Inspur and others (public channel information)

Model Fit MODEL FIT

Use cases128GB HBM2E capacity guidance (estimated: weights × precision + KV/framework overhead)
70B-class inferenceFP8 Quantization fits full model; FP16 full model is close to the limit (estimated)
Hundred-billion-class quantizationINT4/FP4 quantization fits on a single card (estimated)
Cluster TrainingEthernet RoCE scale-out—cluster training on a general-purpose network stack (architectural perspective)
Ecosystem CaveatThe SynapseAI stack is not structurally equivalent to CUDA—assess migration costs per project (ecosystem perspective)
128GB cluster-grade compute — call on demand via the model marketplace
AIXX Cloud · Model Marketplace →

Limitations and Boundaries LIMITS

  • Ecosystem Gap : under CUDA's dominance, SynapseAI's support surface and community resources are smaller (ecosystem perspective)—factor software stack maturity into selection decisions.
  • Roadmap Uncertainty : after Falcon Shores' cancellation, the roadmap follows Intel's current statements (public reporting)—long-term investment requires evaluation.
  • Power Threshold : OAM 900W—plan power/cooling to OAM standards in advance (deployment perspective).

FAQ FAQ

How does Gaudi 3 compare against the H100?
Per Intel's release statements: FP8 1835 TFLOPS (H100 roughly 1979 on the same basis · official page) is close, 128GB HBM2e memory exceeds the H100's 80GB, and interconnect uses 24× 200GbE Ethernet RDMA (the H100 uses NVLink)—Ethernet scale-out is Gaudi's differentiating core; actual throughput varies with software stack maturity (media benchmark results vary).
What's the thinking behind Gaudi 3's Ethernet interconnect?
Official statements: on-card 24× 200GbE—scale-out via standard Ethernet (RoCE/RDMA), no proprietary interconnect hardware, common network stack at scale—this is a distinct path from NVLink's proprietary high-speed domain; the ecosystem is compatible with standard Ethernet operations (architectural perspective).
What's the current state of the Gaudi line?
Public reporting: in early 2025 Intel canceled the successor Falcon Shores discrete product plan and shifted its data center AI strategy—Gaudi 3 is the line's third-generation discrete product, on sale through OEM channels (Dell/Supermicro, etc.); verify current supply and software stack roadmap before purchasing.

References REFERENCES

  • Intel: Gaudi 3 launch release and product page (2024-04 · official statements verified 2026-11) · Original
  • Intel: Habana Gaudi product line documentation · Original
This database
Editorial PolicyAbout this database (disclosure of interests)Errata Channel
Editorial Constitution Summary
Every claim citedFlaws must be stated · Corrections publicly loggedNo paid listings acceptedZero traffic-funneling or sales elements within entriesMarket data mounted only as a data module
AIXX Family
AIXX CloudAIXX ExchangeAIXX MarketAIXX News
Market Roundup: AIXX Exchange · © 2026 AIXX Pedia · Produced by AIXX ·