// RUN ON CONSUMER GPU

Run (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2) on Consumer GPU hardware

33 options ranked by estimated speed at Q4 / 8k context.

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~15,048.6 yes 425k ~$1,999
NVIDIA GeForce RTX 4090 → ~8,464.9 yes 312k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~8,464.9 yes 312k ~$1,999
NVIDIA GeForce RTX 5080 → ~8,061.8 yes 200k ~$999
NVIDIA GeForce RTX 3090 → ~7,860.2 yes 312k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~7,658.7 yes 144k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~7,524.3 yes 200k ~$749
AMD Radeon RX 7900 XTX → ~7,135.1 yes 312k ~$999
NVIDIA GeForce RTX 3080 10GB → ~6,382.2 yes 116k ~$699
NVIDIA GeForce RTX 4080 → ~6,180.7 yes 200k ~$1,199
AMD Radeon RX 7900 XT → ~5,945.9 yes 256k ~$899
NVIDIA GeForce RTX 5070 → ~5,643.2 yes 144k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~5,643.2 yes 200k ~$799
NVIDIA GeForce RTX 3070 Ti → ~5,105.8 yes 88k ~$599
AMD Radeon RX 9070 XT → ~4,786.5 yes 200k ~$599
AMD Radeon RX 7800 XT → ~4,637.8 yes 200k ~$499
NVIDIA GeForce RTX 4070 Super → ~4,232.4 yes 144k ~$599
NVIDIA GeForce RTX 4070 → ~4,232.4 yes 144k ~$599
Intel Arc Pro B70 → ~4,225.5 yes 425k
Intel Arc A770 16GB → ~3,891.9 yes 200k ~$329
AMD Radeon RX 6800 XT → ~3,805.4 yes 200k ~$649
AMD Radeon RX 6900 XT → ~3,805.4 yes 200k ~$999
NVIDIA GeForce RTX 3070 → ~3,762.2 yes 88k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~3,762.2 yes 200k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~3,762.2 yes 88k ~$379
Intel Arc Pro B60 → ~3,169.1 yes 312k
Intel Arc B580 12GB → ~3,169.1 yes 144k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~2,418.5 yes 88k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~2,418.5 yes 200k ~$499
NVIDIA DGX Spark → ~2,292.6 yes 1733k ~$3,999
AMD Ryzen AI Max+ 395 → ~1,902.7 yes 1733k ~$1,999
Intel Arc Pro B50 → ~1,556.8 yes 200k
NVIDIA GeForce RTX 2050 Mobile → ~940.5 yes 32k

Estimate at Q4 (memory-bound). Click any option for its full per-hardware estimate and measured numbers.

FAQ

What Consumer GPU hardware runs (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2)?

On GenAIList, 33 Consumer GPU options can run (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2) at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~15,048.6 tok/s.

How fast is (ensemble): AWD-LSTM-DOC (fin) × 5 (WT2) on Consumer GPU hardware?

Estimated generation speed ranges from ~940.5 to ~15,048.6 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.