// RUN ON CONSUMER GPU

Run iGPT-L 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 → ~2,044.1 yes 415k ~$1,999
NVIDIA GeForce RTX 4090 → ~1,149.8 yes 302k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~1,149.8 yes 302k ~$1,999
NVIDIA GeForce RTX 5080 → ~1,095 yes 190k ~$999
NVIDIA GeForce RTX 3090 → ~1,067.7 yes 302k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~1,040.3 yes 134k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~1,022 yes 190k ~$749
AMD Radeon RX 7900 XTX → ~969.2 yes 302k ~$999
NVIDIA GeForce RTX 3080 10GB → ~866.9 yes 106k ~$699
NVIDIA GeForce RTX 4080 → ~839.5 yes 190k ~$1,199
AMD Radeon RX 7900 XT → ~807.6 yes 246k ~$899
NVIDIA GeForce RTX 5070 → ~766.5 yes 134k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~766.5 yes 190k ~$799
NVIDIA GeForce RTX 3070 Ti → ~693.5 yes 78k ~$599
AMD Radeon RX 9070 XT → ~650.1 yes 190k ~$599
AMD Radeon RX 7800 XT → ~630 yes 190k ~$499
NVIDIA GeForce RTX 4070 Super → ~574.9 yes 134k ~$599
NVIDIA GeForce RTX 4070 → ~574.9 yes 134k ~$599
Intel Arc Pro B70 → ~573.9 yes 415k —
Intel Arc A770 16GB → ~528.6 yes 190k ~$329
AMD Radeon RX 6800 XT → ~516.9 yes 190k ~$649
AMD Radeon RX 6900 XT → ~516.9 yes 190k ~$999
NVIDIA GeForce RTX 3070 → ~511 yes 78k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~511 yes 190k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~511 yes 78k ~$379
Intel Arc Pro B60 → ~430.5 yes 302k —
Intel Arc B580 12GB → ~430.5 yes 134k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~328.5 yes 78k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~328.5 yes 190k ~$499
NVIDIA DGX Spark → ~311.4 yes 1723k ~$3,999
AMD Ryzen AI Max+ 395 → ~258.4 yes 1723k ~$1,999
Intel Arc Pro B50 → ~211.5 yes 190k —
NVIDIA GeForce RTX 2050 Mobile → ~127.8 yes 22k —

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

→ iGPT-L overview → All hardware

FAQ

What Consumer GPU hardware runs iGPT-L?▶

On GenAIList, 33 Consumer GPU options can run iGPT-L at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~2,044.1 tok/s.

How fast is iGPT-L on Consumer GPU hardware?▶

Estimated generation speed ranges from ~127.8 to ~2,044.1 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.