AWS EC2 (H100) — 1024GB / 8TB / 8×GPU
A cloud instance built on the NVIDIA H100 80GB with about 640GB of memory usable for AI — here’s what it runs and how it fits a private AI Business OS.
Some details here are provisional (to-verify). Treat specs as approximate and verify against the manufacturer before relying on them or purchasing.
Configuration
- AI memory
- ~640GB
- Usable for models
- ~563.2GB
- System RAM
- 1024GB
- Storage
- 8TB
- GPUs
- 8×
- Chip
- NVIDIA H100 80GB
Compatible models
Open models graded for this exact configuration.
- DeepSeek-R1 671B (MoE)DeepSeek · ~671B · 128K ctx · MIT
Fits at Q4_K_M (~400GB) with ~163.2GB headroom — about 1 concurrent instance.
Q4_K_M · ~400GBRuns well - Llama 3.1 405BLlama · ~405B · 128K ctx · Llama Community License
Fits at Q8_0 (~410GB) with ~153.2GB headroom — about 1 concurrent instance.
Q8_0 · ~410GBRuns well - Qwen3 235B-A22B (MoE)Qwen · ~235B · 128K ctx · Apache-2.0
Fits at FP16 (~470GB) with ~93.2GB headroom — about 1 concurrent instance.
FP16 · ~470GBRuns well - Qwen2.5 72BQwen · ~72B · 128K ctx · Qwen License
Fits at FP16 (~145GB) with ~418.2GB headroom — about 3 concurrent instances.
FP16 · ~145GBRuns well - Llama 3.1 70BLlama · ~70B · 128K ctx · Llama Community License
Fits at FP16 (~140GB) with ~423.2GB headroom — about 4 concurrent instances.
FP16 · ~140GBRuns well - Llama 3.3 70BLlama · ~70B · 128K ctx · Llama Community License
Fits at FP16 (~140GB) with ~423.2GB headroom — about 4 concurrent instances.
FP16 · ~140GBRuns well - DeepSeek-R1 Distill Llama 70BDeepSeek · ~70B · 128K ctx · MIT
Fits at FP16 (~140GB) with ~423.2GB headroom — about 4 concurrent instances.
FP16 · ~140GBRuns well - Mixtral 8x7B (MoE)Mistral · ~47B · 32K ctx · Apache-2.0
Fits at FP16 (~90GB) with ~473.2GB headroom — about 6 concurrent instances.
FP16 · ~90GBRuns well
Offers
No offers yet. Pricing will appear here once a retailer/affiliate feed is connected.
Turn the AWS EC2 (H100) into a private AI Business OS
Run your own AI agents on hardware you control — private by design, no per-seat data leaving your premises. BrainOutput helps you pick the right machine and turn it into a working AI Business OS.