llamaperf

RTX 2080 Ti vs RTX 5090

For running local LLMs · 66 reports across 12 models

Side A
RTX 2080 Ti
Vendor
nvidia
VRAM
11GB
Memory
Discrete
Side B
RTX 5090
Vendor
nvidia
VRAM
32GB
Memory
Discrete

Highest reported generation speed by model family

These maxima come from different community setups. Model size, quantization, GPU count, offloading and concurrency may differ. They do not establish which GPU is faster for the same workload. Check the comparison conditions →

ModelRTX 2080 TiRTX 5090
Qwen3.8up to 2400B45.0n=1220.0n=34
DeepSeek V4 Flashup to 285B120.0n=10
Qwen3.6up to 35B215.1n=6
Gemma 4up to 31B578.0n=4
Museup to 30B253.0n=4
DeepSeek V4up to 284Bn=1
GLM-5.2up to 744Bn=1
GLM-5.324.5n=1
Kimi K2.5n=1
Qwen2.5up to 32Bn=1
Qwen3up to 235B159.0n=1
Qwen3.5up to 122Bn=1

More comparisons