RTX 3090 vs RTX 4090 for local LLMs

	RTX 3090	RTX 4090
VRAM	24 GB	24 GB
Memory bandwidth	936 GB/s	1008 GB/s
Lowest new price DE	mostly used now	≈€3,955

Measured decode speed, 2026 models (tok/s)

Model	RTX 3090	RTX 4090	Difference
Gemma 4 12B	–	98.1
Gemma 4 26B-A4B	–	164
Gemma 4 31B	–	43.8
LFM2.5 24B-A2B	–	261-339
Nemotron 3.5 Bolt 74B-A9B	–	108
Qwen3.5 9B	–	137
Qwen3.6 27B	40.2	47.0	+17%
Qwen3.6 35B-A3B	146-153	–
Qwen3.8 27B	38.2-71.6	–
gpt-oss 20B	–	203

Sources: RTX 3090 results, RTX 4090 results (community llama-bench via llamabench.ai; quants per result may differ, so treat differences as indicative). Specs per llamabench.ai; prices billiger.de 2026-10-16.

Not sure which you need? Check fit in the VRAM calculator or read best GPU for local LLMs in 2026.

Not enough VRAM? Rent before you buy. Run the same model on a RTX 4090 on a rented 24–96 GB cloud GPU by the hour and see the real speed before spending €1,000+. Try it on RunPod
Referral link: RigBase may earn RunPod credit if you sign up. It never affects our data.

SPONSOR
Your brand here. One sponsor slot on this page, shown to people choosing AI hardware. Sponsor RigBase →