DiggTop ⌕

RTX 3090 x2 vs RTX 3090 x8

Pick a different pair →

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 x27 sourced runs · best 240.8 tok/s
RTX 3090 x84 verified runs · best 91.2 tok/s

Specs and evidence

RTX 3090 x2RTX 3090 x8
VRAM24 GB24 GB
Sourced runs on the board74
Models measured63
Best output tok/s240.8 tok/s91.2 tok/s
Highest measured peak VRAM42.9 GB184.3 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.

The measured matrix

ModelRTX 3090 x2RTX 3090 x8
Qwen3.6-27B240.8 AutoRound INT4—
Qwen3.8-27B219.8 AutoRound INT4 + fp8 KV—
NVIDIA-Nemotron-3.5-Lightning-30B-A3B197.0 NVFP4—
MiniCPM-V 4.7-35B-A3B172.2 int4—
DeepSeek-V4-Flash-0731—91.2 FP8_E4M3
GLM-5.3-Flash—75.2 AWQ-4bit
keys-DeepSeekV4Flash-Vision-EXP-ablit—66.8 FP8
Qwen3.8-Flash-Next-REAP-256-duo53.5 Unsloth-Dynamic-Q3_K_XL—
Qwen3.8-Flash-Next40.1 Unsloth-Dynamic-IQ3_XXS—

9 rows · 6 models on RTX 3090 x2, 3 on RTX 3090 x8 — no verified overlap yet, each side shows its own bests · data as of Oct 17, 2026. Full rules: methodology.

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

↑