// RUN ON CONSUMER GPU

Run Persimmon-8B on Consumer GPU hardware

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

HardwareEst tok/sFits 8k?Max ctx~Price
NVIDIA GeForce RTX 5090 → ~299.4 yes 173k ~$1,999
NVIDIA GeForce RTX 4090 → ~168.4 yes 117k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~168.4 yes 117k ~$1,999
NVIDIA GeForce RTX 5080 → ~160.4 yes 61k ~$999
NVIDIA GeForce RTX 3090 → ~156.4 yes 117k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~152.4 yes 33k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~149.7 yes 61k ~$749
AMD Radeon RX 7900 XTX → ~141.9 yes 117k ~$999
NVIDIA GeForce RTX 3080 10GB → ~127 yes 19k ~$699
NVIDIA GeForce RTX 4080 → ~122.9 yes 61k ~$1,199
AMD Radeon RX 7900 XT → ~118.3 yes 89k ~$899
NVIDIA GeForce RTX 5070 → ~112.3 yes 33k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~112.3 yes 61k ~$799
NVIDIA GeForce RTX 3070 Ti → ~101.6 tight 5k ~$599
AMD Radeon RX 9070 XT → ~95.2 yes 61k ~$599
AMD Radeon RX 7800 XT → ~92.3 yes 61k ~$499
NVIDIA GeForce RTX 4070 Super → ~84.2 yes 33k ~$599
NVIDIA GeForce RTX 4070 → ~84.2 yes 33k ~$599
Intel Arc Pro B70 → ~84.1 yes 173k —
Intel Arc A770 16GB → ~77.4 yes 61k ~$329
AMD Radeon RX 6800 XT → ~75.7 yes 61k ~$649
AMD Radeon RX 6900 XT → ~75.7 yes 61k ~$999
NVIDIA GeForce RTX 3070 → ~74.8 tight 5k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~74.8 yes 61k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~74.8 tight 5k ~$379
Intel Arc Pro B60 → ~63 yes 117k —
Intel Arc B580 12GB → ~63 yes 33k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~48.1 tight 5k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~48.1 yes 61k ~$499
NVIDIA DGX Spark → ~45.6 yes 828k ~$3,999
AMD Ryzen AI Max+ 395 → ~37.8 yes 828k ~$1,999
Intel Arc Pro B50 → ~31 yes 61k —

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

→ Persimmon-8B overview → All hardware

FAQ

What Consumer GPU hardware runs Persimmon-8B?▶

On GenAIList, 32 Consumer GPU options can run Persimmon-8B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~299.4 tok/s.

How fast is Persimmon-8B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~31 to ~299.4 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.