Brand: Gigabyte | Category: GPUs
SKU: GV-NA30-24G | Part #: GV-NA30-24G | MPN: GV-NA30-24G
Contact for Pricing — Request a Quote
The Gigabyte NVIDIA A30 24GB HBM2 PCIe Gen4 Passive GPU Card delivers 24 GB of HBM2 memory with 933 GB/s bandwidth, positioning it as a high-capacity accelerator for memory-intensive enterprise workloads. Built on the NVIDIA Ampere architecture, the A30 combines 3584 CUDA cores and 224 third-generation Tensor cores to achieve 10.3 TFLOPS of FP32 performance, 82 TFLOPS of TF32 Tensor core performance (165 TFLOPS with sparsity), 165 TFLOPS of FP16 Tensor core performance (330 TFLOPS with sparsity), and 330 TOPS of INT8 performance (661 TOPS with sparsity). The 3072-bit memory interface ensures data moves efficiently across the GPU fabric, while passive cooling and a full-height double-slot form factor integrate seamlessly into dense server environments with a modest 165 W thermal design power.
Infrastructure teams deploying AI inference, scientific computing, and virtualization solutions choose Gigabyte NVIDIA A30 cards for their Multi-Instance GPU support—enabling up to four isolated MIG instances per card—and full ECC memory protection for mission-critical applications. NVLink connectivity with support for the NVLink Bridge and 200 GB/s bidirectional throughput allows clustering of multiple GPUs for scaled training and analytics. The PCIe Gen4 x16 interface (part number GV-NA30-24G) provides modern interconnect bandwidth while maintaining backward compatibility with enterprise platforms. An AI infrastructure team or virtualization architect will appreciate the blend of memory capacity, Tensor performance, and architectural flexibility that Gigabyte engineered into this offering.
Omnixon Global supplies the Gigabyte NVIDIA A30 24GB HBM2 PCIe Gen4 Passive GPU Card (GV-NA30-24G) to regional and multinational enterprises across the UAE and GCC. Contact our technical sales team to request a quotation, discuss volume licensing, and explore deployment options tailored to your data center strategy.
| Brand | Gigabyte |
| Category | GPUs |
| SKU | GV-NA30-24G |
| Part Number | GV-NA30-24G |
| Condition | New |
| Manufacturer Part Number | GV-NA30-24G |
| GPU Architecture | NVIDIA Ampere |
| GPU Model | NVIDIA A30 |
| Memory Capacity | 24 GB HBM2 |
| Memory Bandwidth | 933 GB/s |
| Memory Interface Width | 3072-bit |
| CUDA Cores | 3584 |
| Tensor Cores | 224 (3rd Generation) |
| FP32 Performance | 10.3 TFLOPS |
| TF32 Tensor Core Performance | 82 TFLOPS (with sparsity: 165 TFLOPS) |
| FP16 Tensor Core Performance | 165 TFLOPS (with sparsity: 330 TFLOPS) |
| INT8 Tensor Core Performance | 330 TOPS (with sparsity: 661 TOPS) |
| PCIe Interface | PCIe Gen4 x16 |
| Form Factor | Passive cooling, full-height double-slot |
| Thermal Design Power (TDP) | 165 W |
| Display Outputs | None |
| Multi-Instance GPU (MIG) | Supported — up to 4 MIG instances |
| NVLink | Supported (NVLink Bridge, 2 NVLink lanes, 200 GB/s bidirectional) |
| ECC Memory | Supported |
| Server Compatibility | 1U / 2U forced-air datacenter servers |
Reference servers include Dell PowerEdge XE9680 / XE9712, HPE Cray XD670, Lenovo ThinkSystem SR685a / SR675 V3, Supermicro AS-A21GE / SYS-821GE, Gigabyte G593 / G894, ASUS ESC. Share your target platform in the RFQ and we will confirm chassis-to-GPU compatibility and recommended NIC pairing.
Highly model-dependent. L40S / RTX-class: typically 3-6 weeks. H100/H200/B200 in SXM form factor: 12-16 weeks for whole-platform allocations. We quote genuine-channel ETAs only — no grey-market promises.
Yes — Omnixon stocks the full NVIDIA networking lineup (Quantum-2 / Quantum-X InfiniBand, Spectrum-X Ethernet, ConnectX NICs, BlueField DPUs) so we can quote a complete training-cluster BOM, not just the GPUs.
Yes. We hold genuine-channels for NVIDIA AI Enterprise software subscriptions. Add it to your RFQ and we quote node-aligned licensing along with the hardware.