// RUN ON CONSUMER GPU

Run ChatGLM3-6B 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 → ~464 yes 188k ~$1,999
NVIDIA GeForce RTX 4090 → ~261 yes 131k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~261 yes 131k ~$1,999
NVIDIA GeForce RTX 5080 → ~248.6 yes 75k ~$999
NVIDIA GeForce RTX 3090 → ~242.4 yes 131k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~236.1 yes 47k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~232 yes 75k ~$749
AMD Radeon RX 7900 XTX → ~220 yes 131k ~$999
NVIDIA GeForce RTX 3080 10GB → ~196.8 yes 33k ~$699
NVIDIA GeForce RTX 4080 → ~190.6 yes 75k ~$1,199
AMD Radeon RX 7900 XT → ~183.3 yes 103k ~$899
NVIDIA GeForce RTX 5070 → ~174 yes 47k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~174 yes 75k ~$799
NVIDIA GeForce RTX 3070 Ti → ~157.4 yes 19k ~$599
AMD Radeon RX 9070 XT → ~147.6 yes 75k ~$599
AMD Radeon RX 7800 XT → ~143 yes 75k ~$499
NVIDIA GeForce RTX 4070 Super → ~130.5 yes 47k ~$599
NVIDIA GeForce RTX 4070 → ~130.5 yes 47k ~$599
Intel Arc Pro B70 → ~130.3 yes 188k —
Intel Arc A770 16GB → ~120 yes 75k ~$329
AMD Radeon RX 6800 XT → ~117.3 yes 75k ~$649
AMD Radeon RX 6900 XT → ~117.3 yes 75k ~$999
NVIDIA GeForce RTX 3070 → ~116 yes 19k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~116 yes 75k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~116 yes 19k ~$379
Intel Arc Pro B60 → ~97.7 yes 131k —
Intel Arc B580 12GB → ~97.7 yes 47k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~74.6 yes 19k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~74.6 yes 75k ~$499
NVIDIA DGX Spark → ~70.7 yes 842k ~$3,999
AMD Ryzen AI Max+ 395 → ~58.7 yes 842k ~$1,999
Intel Arc Pro B50 → ~48 yes 75k —

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

→ ChatGLM3-6B overview → All hardware

FAQ

What Consumer GPU hardware runs ChatGLM3-6B?▶

On GenAIList, 32 Consumer GPU options can run ChatGLM3-6B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~464 tok/s.

How fast is ChatGLM3-6B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~48 to ~464 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.