Skip to content
Compute
// AGENTIC

Agentic AI

Long-horizon reasoning and tool-use workloads. 288 GB HBM3e holds large KV state and multi-agent context in-memory.

memory

288 GB HBM3e

compute

15 PFLOPS FP4

best for

reasoning + tools

// RECOMMENDED SILICON
// FLAGSHIP

B300

Available

288 GB HBM3e. 15 PFLOPS dense FP4. Built for agentic AI.

MEM
288 GB HBM3e
BW
8 TB/s
NVLINK
Gen 5
$7.95/GPU-hrSpec →
// FLAGSHIP

GB300 NVL72

Available

1.1 ExaFLOPS dense FP4 per rack. The AI-factory standard.

MEM
20.7 TB HBM3e / rack (72 × 288 GB)
BW
8 TB/s
NVLINK
Gen 5
$408,000/moSpec →
// STACK
TensorRT-LLMNVIDIA Dynamospeculative decoding
LaunchBrowse catalog