Intel Gaudi 3 OAM (HL-325L)
OAM (Open Accelerator Module) form factor of Intel's Gaudi 3 with a 900 W power envelope and 24 integrated 200 GbE ports. Used in 8-accelerator HLS-Gaudi 3 baseboards as a Hopper-class alternative without proprietary interconnect.
VRAM
128 GB
Memory
HBM2e
Bandwidth
3700 GB/s
TDP
900W
Suitable Workloads
Large Language Models
Training and inference for models like GPT-4, Llama 70B+
Deep Learning Training
High-performance training for neural networks
Distributed Training
Multi-node training with fast interconnects
High-Throughput Inference
Optimized for batched inference workloads
Key Highlights
- Built on Habana Gaudi 3 architecture
- 8 AI accelerator cores for tensor operations
- 64 compute cores for parallel processing
- OAM form factor
Memory
Compute Performance
Architecture
Power & Physical
Interconnect
Pricing
Notes
compute_cores reflects 64 TPC (Tensor Processor Cores); ai_accelerators is the count of dedicated MME (Matrix Multiplication Engines). Compute figures are dense (no sparsity).
Models That May Fit on Intel Gaudi 3 OAM (HL-325L)
Estimates based on INT8 quantization. Actual fit depends on framework and batch size.
Explore More
Added Apr 30, 2026
Last updated: Apr 30, 2026
Start building with the right model.
Automatically route workloads to the right model for every task, every time.