GPU infrastructure planning for production
Capacity modeling for teams deploying LLMs and diffusion models at scale.
Used by engineering teams from
Design clusters for peak throughput
Model KV cache growth, concurrency limits, and multi-node latency before you provision a single GPU.
Every frontier model, updated with each new release
Evaluate the latest open-weight models and quantization footprints across datacenter GPUs, RTX workstations, and Apple Silicon.
Calibrated on physical GPU telemetry
Every memory and throughput curve is continuously calibrated against live runs across NVIDIA, AMD, and Apple Silicon hardware.
Plan clusters with measured hardware telemetry before provisioning compute.
Open-weight model inference specifications
Context windows, quantization footprints, and target GPUs across leading open-weight architectures.
| Model | Parameters | Attention | Context | Min VRAM | Runs on |
|---|---|---|---|---|---|
| Kimi K3.1 Pro Moonshot AI | 2800B A104B | DSA | 1.0M | 1321.7 GB | 32x H100 |
| Qwen3.8 2.4T A95B Alibaba | 2400B A95B | GQA | 262.1k | 1132.7 GB | 16x H100 |
| GLM-5.3 Z.ai | 753.9B A51.6B | DSA | 1.0M | 356.4 GB | 8x H100 |
| DeepSeek V4.1 Flash DeepSeek | 552B A16B | CSA2 | 1.0M | 261.1 GB | 4x H100 |
| DeepSeek-V4-Flash-Vision-Exp DeepSeek | 290.9B A20.4B | DSA | 1.0M | 138.0 GB | 2x H100 |
| K2 Horizon MoVA 36B A4B Moonshot AI | 36B A4B | GQA | 524.3k | 17.8 GB | 1x RTX 4090 |
| Qwen 3.8 27B Alibaba | 27B | GQA | 262.1k | 13.2 GB | 1x 16GB GPU |
| K2 Horizon 7B Moonshot AI | 7B | GQA | 524.3k | 3.9 GB | 1x 8GB GPU |
GPU & workstation profiles
Explore hardware directory →NVIDIA Blackwell B300 (288GB)
288 GB · 8000 GB/s
NVIDIA B200 (192GB)
192 GB · 8000 GB/s
Instinct MI325X (256GB)
256 GB · 6000 GB/s
Instinct MI300X (192GB)
192 GB · 5300 GB/s
NVIDIA H100 (80GB)
80 GB · 3350 GB/s
NVIDIA RTX 5090 (32GB)
32 GB · 1792 GB/s
Apple M5 Max (128GB)
96 GB · 614 GB/s
NVIDIA RTX 4090 (24GB)
24 GB · 1008 GB/s
Engineering courses & roadmaps
150+ courses covering transformer internals, quantization kernels, distributed inference serving, and PyTorch foundations.
Hardware benchmarks & deployment guides
GPU architecture comparisons, model sizing breakdowns, and serving optimization guides.