ByteDance SeedVR2 distilled to 1.44B, cutting RAM to 4.6 GB for local image upscaling
TimeTruth2490 · reddit · 2026-07-30
A researcher reports distilling ByteDance’s SeedVR2-7B image upscaler into a much smaller 1.44B, 6-layer model called SeedVR2-1.4B.
Highlights:
- The student is about 5.7× smaller on disk and needs about 4.6 GB of RAM versus roughly 14–16 GB for the teacher.
- On an Apple M2 Ultra, it is around 1.6× faster at 4× upscaling and 4.7–5.6× faster at 8× upscaling.
- Quality is measured against the teacher with band-energy metrics and MAE across 15 scenes.
- The author concludes it is best used as a 2×–4× upscaler; 8× works but clearly degrades.
- The model is published on Hugging Face.
More from Infra
- NVIDIA launches RTX Spark for local AI on Windows PCs — ostrisai · 2026-10-16
- SkyRL ships Day-0 RL support for Beam with +62% LoRA decode throughput in vLLM — vllm_project · 2026-10-16
- DFlash2 delivers up to 3.03x decode-only speedup for Beam across five benchmarks — vllm_project · 2026-10-16
- vLLM's Beam serving kernels cut GB200 forward time from 8.15 to 6.47 ms at TP8 batch 128 — vllm_project · 2026-10-16
- Reflection's 501B MoE Beam hits 463 tok/s on 8x GB200 with NVFP4 at day 0 — zhyncs42 · 2026-10-16
- Gemma 4 31B QAT on one 24GB card: 46 tok/s plain, ~80 with draft model on RTX 4090 — Subject_Poet_3861 · 2026-10-16