Flopper.io GPU Specification Sheet
September 30, 2026
flopper.io/gpu/nvidia-groq-3-lpuNVIDIA Groq 3 LPU
Groq LPU 2026
Overview
—
FP32 TFLOPS
—
Memory On-die SRAM
—
Memory Bandwidth
—
TDP
Performance Metrics
| Precision | Peak (dense) | Bits | Type |
|---|---|---|---|
| FP64 | — | — | Standard |
| FP32 | — | — | Standard |
| TF32 | — | — | Standard |
| BF16 | — | — | Standard |
| FP16 | — | — | Standard |
| FP8 | 1200.0 | 8 | Standard |
| FP6 | — | — | Standard |
| FP4 | — | — | Standard |
| INT8 | — | — | Standard |
Memory Specifications
- Capacity
- —
- Type
- On-die SRAM
- Bandwidth
- —
- Interface Width
- —
Power & Efficiency
- TDP
- —
- Max power
- —
- FP32 dense TFLOPS per W of TDP
- —
- FP16 dense TFLOPS per W of TDP
- —
Per W of TDP figures divide dense peak by accelerator TDP. Board power only: excludes host CPUs, networking, cooling and facility overhead.
Hardware Details
Vendor: NVIDIA
Architecture: Groq LPU
Process Node: —
Form Factor: —
Launch Year: 2026
Interconnect
GPU-to-GPU: LPU C2C
Bandwidth: 2.5 TB/s direction and scope not stated by vendor