// RUN ON CONSUMER GPU

Run DeciLM 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 → ~488.4 yes 189k ~$1,999
NVIDIA GeForce RTX 4090 → ~274.7 yes 133k ~$1,599
NVIDIA GeForce RTX 3090 Ti → ~274.7 yes 133k ~$1,999
NVIDIA GeForce RTX 5080 → ~261.7 yes 77k ~$999
NVIDIA GeForce RTX 3090 → ~255.1 yes 133k ~$1,499
NVIDIA GeForce RTX 3080 Ti → ~248.6 yes 48k ~$1,199
NVIDIA GeForce RTX 5070 Ti → ~244.2 yes 77k ~$749
AMD Radeon RX 7900 XTX → ~231.6 yes 133k ~$999
NVIDIA GeForce RTX 3080 10GB → ~207.1 yes 34k ~$699
NVIDIA GeForce RTX 4080 → ~200.6 yes 77k ~$1,199
AMD Radeon RX 7900 XT → ~193 yes 105k ~$899
NVIDIA GeForce RTX 5070 → ~183.2 yes 48k ~$549
NVIDIA GeForce RTX 4070 Ti Super → ~183.2 yes 77k ~$799
NVIDIA GeForce RTX 3070 Ti → ~165.7 yes 20k ~$599
AMD Radeon RX 9070 XT → ~155.4 yes 77k ~$599
AMD Radeon RX 7800 XT → ~150.5 yes 77k ~$499
NVIDIA GeForce RTX 4070 Super → ~137.4 yes 48k ~$599
NVIDIA GeForce RTX 4070 → ~137.4 yes 48k ~$599
Intel Arc Pro B70 → ~137.1 yes 189k —
Intel Arc A770 16GB → ~126.3 yes 77k ~$329
AMD Radeon RX 6800 XT → ~123.5 yes 77k ~$649
AMD Radeon RX 6900 XT → ~123.5 yes 77k ~$999
NVIDIA GeForce RTX 3070 → ~122.1 yes 20k ~$499
NVIDIA GeForce RTX 5060 Ti 16GB → ~122.1 yes 77k ~$429
NVIDIA GeForce RTX 5060 Ti 8GB → ~122.1 yes 20k ~$379
Intel Arc Pro B60 → ~102.9 yes 133k —
Intel Arc B580 12GB → ~102.9 yes 48k ~$249
NVIDIA GeForce RTX 4060 Ti 8GB → ~78.5 yes 20k ~$399
NVIDIA GeForce RTX 4060 Ti 16GB → ~78.5 yes 77k ~$499
NVIDIA DGX Spark → ~74.4 yes 843k ~$3,999
AMD Ryzen AI Max+ 395 → ~61.8 yes 843k ~$1,999
Intel Arc Pro B50 → ~50.5 yes 77k —

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

→ DeciLM 6B overview → All hardware

FAQ

What Consumer GPU hardware runs DeciLM 6B?▶

On GenAIList, 32 Consumer GPU options can run DeciLM 6B at Q4 — the fastest is the NVIDIA GeForce RTX 5090 at ~488.4 tok/s.

How fast is DeciLM 6B on Consumer GPU hardware?▶

Estimated generation speed ranges from ~50.5 to ~488.4 tok/s at Q4 (memory-bandwidth bound). Click any option for its full estimate and any measured numbers.