Gaudi 2 Specifications
Gaudi 2 supports all popular data types required for deep learning: FP32, TF32, BF16, FP16 & FP8 (both E4M3 and E5M2).
96 GB
HBM2e Memory
Optimized capacity for FP8 models with large context window and batch.
2.4 Tbps
RoCE v2 Bandwidth
Fast GPU interconnect for training and multi-node inference.
865
TFLOPS FP8
Superior token cost and performance at FP8 precision.
2.8X
Faster Inference
vs A00 at FP8 performance of A100, and 1.4x at BF16.
Intel Gaudi 2 on Denvr AI Cloud
Best Price-Performance
96 GB HBM2e at lowest price point. Near-H100 performance on supported models for teams optimizing cost per token.
LLM Training & Inference
Native support for PyTorch and Hugging Face Optimum. Train and serve popular open-weight models including Llama, Mixtral, and Qwen.
Alternate Silicon
Evaluate non-NVIDIA accelerators to maximize your AI compute budget. Train or operate models via vLLM serving engine.
Managed Storage
High-performance Weka filesystem and local NVMe available. No external storage to provision for datasets, checkpoints, or model artifacts.
Platform
GPUs
GPU VRAM
vCPUs
Memory
Local Storage
Interconnect
On-Demand
Intel Gaudi 2
8
96 GB
160
1024 GB
4x 7.6TB NVMe
-
$1.25 / GPU
Configurations
Per-minute billing with on-demand and reserved options. All configurations available as bare metal, VM, or model endpoints.
Related GPUs
Compare Denvr GPU options by workload and performance requirements.
Optimized For
Distributed training, multi-node scaling
Cost-effective inference, fine-tuned models
Large model training, high-throughput inference
VRAM
80 GB
96 GB
80 GB
Memory Bandwidth
2,039 GB/s
2,450 GB/s
3,350 GB/s
FP64/FP32
19.5 TFLOPS
-
67 TFLOPS
FP16
312 TFLOPS
432 TFLOPS
1,979 TFLOPS
FP8
-
865 TFLOPS
3,958 TFLOPS
NVLink
600 GB/s
-
900 GB/s
On-Demand Pricing
$1.30 / GPU
$1.25 / GPU
$2.10 / GPU
Infrastructure you can trust at scale
As an Intel Partner Alliance member we build and operate AI clusters following vendor reference architectures. Your models and data are supported via strict privacy safeguards and SOC 2 Type 2 security practices.
