From the RTX 5060 Ti 16 GB to the RTX 5070 Ti 16 GB: what changes for local AI
+0 GB of VRAM (16 → 16 GB). Of 49 models, 0 newly run well, 0 get a better file or stop offloading, and 49 stay the same.
0newly run well
0better file
49no change
+0GB more VRAM
01No change in what fits
FLUX.1 dev Runs wellFLUX.1 schnell Runs wellFLUX.1 Kontext Runs wellFLUX.1 Krea Runs wellFLUX.1 Fill Runs wellFLUX.2 dev Offload onlyFLUX.2 klein 9B Runs wellFLUX.2 klein 4B Runs wellKrea 2 Runs wellQwen-Image RunsQwen-Image-Edit TightQwen-Image 2.1 Runs wellZ-Image Turbo Runs wellZ-Image Runs wellIdeogram 4 RunsBoogu-Image Runs wellERNIE-Image Runs wellHiDream-O1 Runs wellMage-Flow Runs wellLumina 2.0 Runs wellHiDream-I1 Full RunsHiDream-I1 RunsSD 3.5 Large Runs wellSD 3.5 Medium Runs wellChroma1-HD Runs wellSDXL Runs wellIllustrious / Pony Runs wellSD 1.5 Runs wellHunyuanImage 2.1 RunsWan 2.1 14B RunsWan 2.1 1.3B Runs wellWan 2.1 I2V 480P RunsWan 2.1 I2V 720P TightWan VACE 14B TightWan 2.2 T2V RunsWan 2.2 I2V RunsWan 2.2 5B Runs wellWan 2.2 Animate TightWan Animate 2 TightWan 2.2 S2V TightSCAIL-2 TightHunyuanVideo 13B RunsLTX-Video 13B RunsHunyuanVideo 1.5 Runs wellLTX-2 TightLTX-2.3 TightLTX-2.5 TightMiniMax H3 Offload onlyMiniMax H3 Pruned Tight
Same verdict and same file on both cards. Speed can still differ.
02Speed and features
Memory bandwidth goes from 448 to 896 GB/s (×2.00). For models that fit on both cards, bandwidth and compute decide the speed; this ratio is a rough first guide, not a benchmark.