What is the largest LLM the AMD Radeon RX 7900 XT can run?

Nemotron 3 Nano 30B-A3B, at Q4_K_M using about 19.8 GB of the 20 GB available, at roughly 226 tokens per second (8K context).

How many local AI models run well on the AMD Radeon RX 7900 XT?

41 of the 69 models we track run fully in its 20 GB of VRAM at Q4_K_M or better.

Generation speed follows memory bandwidth. With 800 GB/s, expect about 30 tokens per second on an 8B model.

Every model we track, checked against the AMD Radeon RX 7900 XT at 8K context. The largest that runs well is Nemotron 3 Nano 30B-A3B.

VRAM20 GBAMD

Bandwidth800GB/s

Runs well41/69models at Q4_K_M or better

Fastest model~327tok/s · Qwen3.6 35B-A3B

Bars show memory used out of 20 GB. Pick a row for the full breakdown.

NameMemory usedBest quantSpeed

NameMemory usedBest quantSpeed

NameMemory usedBest quantSpeed

Run it in the cloud

Rent a cloud GPU by the hour. No upfront cost, stop any time.