Flopper.io GPU Comparison Sheet
October 21, 2026
flopper.io/compare/nvidia-b100-sxm-192gb-vs-openai-jalapenoNVIDIA B100 SXM 192GB vs OpenAI Jalapeño
Side by side datacenter GPU specifications from the Flopper.io database.
Overview
| NVIDIA B100 SXM 192GB | OpenAI Jalapeño | |
|---|---|---|
| FP32 TFLOPS | 60.0 | — |
| Memory | 192 GB | 216 GB |
| Memory Bandwidth | 8.0 TB/s | 15.4 TB/s |
| TDP | 700 W | 700 W |
Performance by Precision
| Precision | NVIDIA B100 SXM 192GB peak TFLOPS (dense) | OpenAI Jalapeño peak TFLOPS (dense) |
|---|---|---|
| FP64 | 30.0 | — |
| FP32 | 60.0 | — |
| TF32 | 875.0 | — |
| BF16 | 1750.0 | — |
| FP16 | 1750.0 | — |
| FP8 | 3500.0 | 3400.0 |
| FP6 | — | — |
| FP4 | 7000.0 | 13400.0 |
| INT8 | 3500.0 | — |
Specifications
| NVIDIA B100 SXM 192GB | OpenAI Jalapeño | |
|---|---|---|
| Architecture | Blackwell | Jalapeño |
| Process Node | 4nm | — |
| Launch Year | 2024 | 2026 |
| Form Factor | SXM | Custom |
| Memory Type | — | HBM4 |
| Interface Width | — | — |
| Interconnect | — | Ethernet scale-up (Broadcom Tomahawk 6) |
| TDP | 700 W | 700 W |
| FP32 Efficiency | 0.09 TFLOPS/W | — |