Rent B200 GPUs
on Vast.ai for
The NVIDIA B200, featuring 192GB of HBM3e memory and 16896 x2 cores, is a unified AI platform for develop-to-deploy pipelines for businesses of any size at any stage in their AI journey.
No current offersThe rent link will appear when matching B200 listings are available.
B200-Ready Models
These popular AI models can run on the B200. Actual requirements vary by model, engine, precision, and configuration.
textvision
Kimi K3
Moonshot AI's 2.8T-parameter multimodal MoE — full-precision vLLM flagship on 8xB300, plus single-node Unsloth 1-bit and 2-bit dynamic GGUF quants
text
Flint
Recursive's 40B-parameter coding MoE with about 2.5B active parameters per token and a 262K-token context, built for coding agents
textvision
NVIDIA Nemotron 3.5 Bolt 74B A9B
Hybrid Mamba-Transformer MoE vision-language model with 9B active params, built for local agentic work with tool calling and reasoning control
B200 GPU Details
| Category | B200 |
|---|---|
| GPU Name | GB100 |
| Architecture | Blackwell |
| Process Size | 5 nm |
| Transistors | 104,000 million |
| Release Date | 2024 |
| Base Clock | 1665 MHz |
| Boost Clock | 1837 MHz |
| Memory Size | 96 GB |
| Memory Type | HBM3e |
| Bandwidth | 4.10 TB/s |
| Tensor Cores | 528 |
| FP16 (half) | 248.3 TFLOPS (4:1) |
| FP32 (float) | 62.08 TFLOPS |