TimeLordRaps
/

Idol-GGUF

@@ -40,6 +40,8 @@ GB means one billion bytes; MiB means 1,048,576 bytes. PyTorch allocated memory
 The third-party [DINOv3 Q4 file](https://huggingface.co/zisisnotzis/dinov3-gguf) was tested at revision `7fbb86b66bbb1dd483c94c179d35c969c76467e6`; its SHA-256 is `cf80ee6e3c2d4044b3214c0ccde5e416cb999b799f35b699f4edaa4059d83d71`. It is an experimental compatibility result, **not a quality recommendation**: for the axolotl image and matched staged settings, its latent differed by 34.17% in relative Euclidean norm from the same Idol Q8 run using the original DINOv3 encoder. One image and seed are insufficient to establish general behavior. The FLUX.2 autoencoder remained at original precision, and its separate terms apply.
 ## Evaluation limits
 On one NVIDIA GeForce RTX 4090 workstation, with 50 sampling steps, 262,144 Gaussian points, original-precision image encoders, resident graphics-device models, batched guidance, and the experimental fused and optimized-attention paths, two image reconstructions were compared with original Idol weights. Relative latent error was 3.23% for the lamp and 6.47% for the axolotl. These are agreement measurements, not ground-truth 3D accuracy. In three fixed diagnostic orthographic views of the axolotl, calibrated Q8 had lower foreground color error than an otherwise comparable uncalibrated Q8 candidate; silhouette overlap improved in two of the three views.

 The third-party [DINOv3 Q4 file](https://huggingface.co/zisisnotzis/dinov3-gguf) was tested at revision `7fbb86b66bbb1dd483c94c179d35c969c76467e6`; its SHA-256 is `cf80ee6e3c2d4044b3214c0ccde5e416cb999b799f35b699f4edaa4059d83d71`. It is an experimental compatibility result, **not a quality recommendation**: for the axolotl image and matched staged settings, its latent differed by 34.17% in relative Euclidean norm from the same Idol Q8 run using the original DINOv3 encoder. One image and seed are insufficient to establish general behavior. The FLUX.2 autoencoder remained at original precision, and its separate terms apply.
+A separate Linux container test loaded the released Q8 file and original encoders and served one 50-step image-to-3D request in 64.54 seconds including loading. PyTorch reported 3.51 GB peak allocated GPU memory, excluding the graphics context and other processes. It used a new virtual environment inheriting an existing GPU-enabled base image, with unrelated inherited dependency conflicts; the repository Dockerfile has not yet completed a clean build. This observation is not a latency distribution.
 ## Evaluation limits
 On one NVIDIA GeForce RTX 4090 workstation, with 50 sampling steps, 262,144 Gaussian points, original-precision image encoders, resident graphics-device models, batched guidance, and the experimental fused and optimized-attention paths, two image reconstructions were compared with original Idol weights. Relative latent error was 3.23% for the lamp and 6.47% for the axolotl. These are agreement measurements, not ground-truth 3D accuracy. In three fixed diagnostic orthographic views of the axolotl, calibrated Q8 had lower foreground color error than an otherwise comparable uncalibrated Q8 candidate; silhouette overlap improved in two of the three views.