FitMyLLM

Running LLMs on AMD Radeon Pro WX 9100: What You Can Run

The AMD Radeon Pro WX 9100 packs 16GB of HBM2 with 484 GB/s of memory bandwidth, making it a strong mid-range option for running local AI models. Here is which models fit, how fast they run, and what quantisation gives the best balance of quality and speed.

Already have it? Find the best model for it. Still choosing? See what a budget buys. Or the full specifications.

Loading the model tables…

What determines LLM performance on the AMD Radeon Pro WX 9100?

Two specs decide almost everything when you run a language model locally. Memory capacity decides which models fit at all, and memory bandwidth decides how fast they write, because every generated token reads the whole model once.

With 16GB of VRAM, the AMD Radeon Pro WX 9100 holds models up to roughly 25B parameters at Q4_K_M, the most common quantisation, once the runtime has taken its own share. The context you ask for needs room on top of that, so a model near the limit runs with a short context.

At 484 GB/s, speed falls as models grow: a model twice the size generates at roughly half the tokens per second. The estimates in the tables come from the bandwidth and each model's size, and real runs differ by 10–20% with context length and the inference backend.

Our recommendation

For general chat and everyday use on the AMD Radeon Pro WX 9100, start with a 8B–14B model at Q4_K_M. It is the best balance of answer quality and generation speed this card offers.

For coding, pick a model built for it (“coder” in the name). For reasoning-heavy work, reasoning models such as DeepSeek R1 distills or Qwen with thinking mode do best, at the cost of longer answers.

The AMD Radeon Pro WX 9100 specs page has the full specifications.