Best sourced output tok/s per model (single-stream generation) — engine and version, quantization, context and a source link on every row. Hover a value for the quant.
Radeon AI Pro R9700 x39 sourced runs · best 223.4 tok/s
RTX 5090 x21 verified run · best 195.6 tok/s
Specs and evidence
Radeon AI Pro R9700 x3
RTX 5090 x2
VRAM
32 GB
32 GB
Sourced runs on the board
9
1
Models measured
9
1
Best output tok/s
223.4 tok/s
195.6 tok/s
Highest measured peak VRAM
25.8 GB
57.2 GB
Street price (latest snapshot)
—
—
Sourced rows only. Peak VRAM is recorded by the person who ran the benchmark, so it is a floor for what the card can hold, not a specification. Price snapshots.
10 rows · 9 models on Radeon AI Pro R9700 x3, 1 on RTX 5090 x2 — no verified overlap yet, each side shows its own bests · data as of Oct 17, 2026. Full rules: methodology.