// RUN ON CONSUMER GPU

Run Llama 3.2 11B on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~262.6 yes 134k ~$1,999
NVIDIA GeForce RTX 4090 → ~147.7 yes 89k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~147.7 yes 89k ~$1,999
NVIDIA GeForce RTX 5080 → ~140.7 yes 44k ~$999
NVIDIA GeForce RTX 3090 → ~137.2 yes 89k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~133.7 yes 22k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~131.3 yes 44k ~$749
AMD Radeon RX 7900 XTX → ~124.5 yes 89k ~$999
NVIDIA GeForce RTX 3080 10GB → ~111.4 yes 11k ~$699
NVIDIA GeForce RTX 4080 → ~107.9 yes 44k ~$1,199
AMD Radeon RX 7900 XT → ~103.8 yes 67k ~$899
NVIDIA GeForce RTX 5070 → ~98.5 yes 22k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~98.5 yes 44k ~$799
AMD Radeon RX 9070 XT → ~83.5 yes 44k ~$599
AMD Radeon RX 7800 XT → ~80.9 yes 44k ~$499
NVIDIA GeForce RTX 4070 Super → ~73.9 yes 22k ~$599
NVIDIA GeForce RTX 4070 → ~73.9 yes 22k ~$599
Intel Arc Pro B70 → ~73.7 yes 134k —
Intel Arc A770 16GB → ~67.9 yes 44k ~$329
AMD Radeon RX 6800 XT → ~66.4 yes 44k ~$649
AMD Radeon RX 6900 XT → ~66.4 yes 44k ~$999
NVIDIA GeForce RTX 5060 Ti 16GB → ~65.7 yes 44k ~$429
Intel Arc Pro B60 → ~55.3 yes 89k —
Intel Arc B580 12GB → ~55.3 yes 22k ~$249
NVIDIA GeForce RTX 4060 Ti 16GB → ~42.2 yes 44k ~$499
NVIDIA DGX Spark → ~40 yes 658k ~$3,999
AMD Ryzen AI Max+ 395 → ~33.2 yes 658k ~$1,999
Intel Arc Pro B50 → ~27.2 yes 44k —

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

→ Llama 3.2 11B overview → All hardware

FAQ

What Consumer GPU hardware runs Llama 3.2 11B?▶

On GenAIList, 28 Consumer GPU options can run Llama 3.2 11B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~262.6 tok/s.

How fast is Llama 3.2 11B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~27.2 to ~262.6 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.