RTX 3090 x2 vs RX 7900 XTX

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.

RTX 3090 x28 sourced runs · best 240.8 tok/s

RX 7900 XTX6 verified runs · best 550.2 tok/s

Specs and evidence

	RTX 3090 x2	RX 7900 XTX
VRAM	24 GB	24 GB
Sourced runs on the board	8	6
Models measured	7	5
Best output tok/s	240.8 tok/s	550.2 tok/s
Highest measured peak VRAM	42.9 GB	22.9 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.

Head-to-head

Only one model has been measured on both cards (Qwen3.8-27B). On that single data point RTX 3090 x2 records 219.8 tok/s and RX 7900 XTX records 154.2 tok/s — treat it as one data point, not a ranking.

The measured matrix

Model	RTX 3090 x2	RX 7900 XTX
Qwen3.5-0.8B	—	550.2 MQ4
LFM2.5-350M	—	398.4 MQ4
Qwen3.6-27B	240.8 AutoRound INT4	—
Qwen3.8-27B	219.8 AutoRound INT4 + fp8 KV	154.2 MQ4
NVIDIA-Nemotron-3.5-Lightning-30B-A3B	197.0 NVFP4	—
MiniCPM-V 4.7-35B-A3B	172.2 int4	—
Qwen3.6-35B-A3B	—	150.0 MQ4R
Qwen3.6-35B-A3B-abliterated-v4	—	82.1 Q8_0
Nanbeige4.3-33B	66.2 W4A16 (INT4 AutoRound, GPTQ, gro	—
Qwen3.8-Flash-Next-REAP-256-duo	53.5 Unsloth-Dynamic-Q3_K_XL	—
Qwen3.8-Flash-Next	40.1 Unsloth-Dynamic-IQ3_XXS	—

Head-to-head (models tested on both): RTX 3090 x2 1 · RX 7900 XTX 0 · tied 0 · data as of Oct 30, 2026. Full rules: methodology.

Related reading: RTX 5090 vs 3090 · quantization · Value index · Full board