🔌
New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
// RUN — FITS
Can the NVIDIA H100 SXM5 80GB run OmniFusion-7B (InternViT-6B-448px V1-2)?
Yes — here's how fast and up to what context, with real measured numbers.
OmniFusion-7B (InternViT-6B-448px V1-2) on the NVIDIA H100 SXM5 80GB
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 287k | 97.5-116.2 t/s | ~1.2-1.4 (slow) |
| Q8 | yes | 364k | 195-232.4 t/s | ~2.3-2.8 (slow) |
| Q4 | yes | 397k | 348.2-415 t/s | ~4.2-5 (slow) |
Estimate (memory-bound). Beyond "max context" the KV cache spills to system RAM offload (depends on your system RAM) and speed drops to the "beyond" figure.
→ All models on the NVIDIA H100 SXM5 80GB
→ OmniFusion-7B (InternViT-6B-448px V1-2) specs
→ Other hardware
FAQ
Can the NVIDIA H100 SXM5 80GB run OmniFusion-7B (InternViT-6B-448px V1-2)?▶
Yes. At Q4 it generates ~348.2-415 tok/s and fits up to 397k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~4.2-5 tok/s.
How much context fits?▶
At Q4, up to about 397k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.