New: connect Claude & other AIs to GenAIList over MCP — research the catalog and contribute to the shared knowledge base. Learn how →
Can the NVIDIA GeForce RTX 4080 run TAPE Transformer?
Yes — here's how fast and up to what context, with real measured numbers.
TAPE Transformer on the NVIDIA GeForce RTX 4080
| Quant | Fits? | Max context | Speed | Beyond context |
|---|---|---|---|---|
| FP16 | yes | 201k | 7069.5-8425.3 t/s | ~384.2-457.9 (slow) |
| Q8 | yes | 201k | 14138.9-16850.5 t/s | ~768.4-915.8 (slow) |
| Q4 | yes | 201k | 25248.1-30090.2 t/s | ~1372.2-1635.3 (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.
FAQ
Can the NVIDIA GeForce RTX 4080 run TAPE Transformer?▶
Yes. At Q4 it generates ~25248.1-30090.2 tok/s and fits up to 201k context; beyond that it spills to system RAM offload (depends on your system RAM) and slows to ~1372.2-1635.3 tok/s.
How much context fits?▶
At Q4, up to about 201k tokens stay in fast memory; longer context spills to system RAM offload (depends on your system RAM) and throughput collapses.