Felipe Sztutman

felipesztutman

sztlink
sztlink
felipesztutman

AI & ML interests

I quantize models and hang the results on gallery walls. CUDA kernels, fidelity benchmarks, installations. Building the lowest-level realtime diffusion stack. São Paulo.

Recent Activity

updated a model 18 days ago

felipesztutman/MiniMax-H3-W4A4

updated a model 18 days ago

felipesztutman/MiniMax-H3-W4A4

liked a model 3 months ago

google/gemma-3-12b-it-qat-q4_0-unquantized

View all activity

Organizations

None yet

New activity in krea/krea-realtime-video 3 months ago

Runs on a single RTX 4090 in W4A4 (2.81 fps, open weights and receipts)

❤️ 1

#9 opened 3 months ago by

felipesztutman

New activity in Comfy-Org/Krea-2 3 months ago

The sampling acceleration from nvfp4 quantization in Krea2 is not significant.

#6 opened 4 months ago by

Aca233

MXFP8 for Raw?

#9 opened 4 months ago by

wywywywy