·
AI & ML interests
Agentic AI for self exploration
Recent Activity
reacted to savoji's post with 🔥 7 days ago Idol, a new Apache-2.0 image-to-3D model from @speridlabs, makes a 3D Gaussian splat from one image in 7.2 GB of peak VRAM. It is the #1 trending model on the Hub with 1,057 likes.
It writes a standard 3DGS .ply with up to 1,048,576 Gaussians, including sides the picture never showed, and does background removal and framing itself. The catch: the only quality evidence is the authors' own benchmark, on a repo created 2026-10-28.
📊 The numbers
• Authors' benchmark, 1 x H100: best in 6 of 8 columns vs TripoSplat, TRELLIS and LGM (SSIM 0.726, LPIPS 0.213, DINOv2 0.781, CLIP 0.885)
• LGM wins PSNR (12.46 vs 12.25) and speed (1.1 s). Idol takes 6.8 s vs 7.1 s for TRELLIS and 9.9 s for TripoSplat
• 262,144 Gaussians, 50 steps: 3.4 s on B300, 6.7 s on H100, 19.1 s on RTX 4090, 46.2 s on a 10 GB RTX 3080
🛠️ Try it
• Install idol-sdk, accept the terms of the gated DINOv3 and FLUX.2-dev models, run hf auth login
• python -m idol.sample --image photo.jpg --mode real
• --num_gaussians runs from 32,768 (0.4 MB) to 1,048,576 (12.2 MB)
⚠️ Caveats
• Test set and size are not named, and no independent evaluation exists yet
• The card says "under 2 seconds", but the timing table shows 3.4 s at best
• Single objects only, NVIDIA CUDA required, no failure cases shown
https://huggingface.co/speridlabs/idol
https://speridlabs.com/research/idol
If you run it on your own messy photos, how do the unseen sides hold up?
📚 Read it elsewhere
• The deep dive on Medium: https://medium.com/p/5f660ae6ae56
• The long read on Substack: https://sajjadpakdamansavoji.substack.com/p/speridlabs-idol-builds-3d-splats
• The thread on X: https://x.com/saavoji/status/2119137739580858545
• Every version on sajjadpsavoji.github.io: https://sajjadpsavoji.github.io/blog#speridlabs-idol View all activity Organizations
8B • Updated • 1.38k
• 18
Image-Text-to-Text • 5B • Updated • 27
• 7
Sweaterdog/andy-4-micro-safetensors-0516
2B • Updated • 70
Sweaterdog/GRaPE-Mini-Beta-Thinking
Text Generation • 2B • Updated • 126
• 5
Sweaterdog/GRaPE-Mini-Beta
Text Generation • 2B • Updated • 219
• 4
Text Generation • 0.4B • Updated • 73
• 2
Sweaterdog/Andy-4-0516-Safetensors
Updated • 14
• 3
Sweaterdog/andy-4-micro-base
2B • Updated • 4
8B • Updated • 9
Sweaterdog/Andy-4-tiny-safetensors
0.4B • Updated • 21
• 1
Sweaterdog/Andy-4-base1-LoRA
Updated
Sweaterdog/Andy-4-base-LoRA-DEPRECATED
Updated
Sweaterdog/Andy-4-base-DEPRECATED
8B • Updated • 809
• 2
2B • Updated • 219
• 4
Sweaterdog/Andy-4-tiny-base-LoRA
Updated
Sweaterdog/Andy-4-tiny-LoRA
Updated
Sweaterdog/Andy-4-tiny-LoRA-temp
Updated
Sweaterdog/Smol-reason2.1-safetensors
Sweaterdog/Andy-4-0516-LoRA
Updated
Sweaterdog/Andy-4-micro-0516-LoRA
Sweaterdog/Andy-4-micro-0427
2B • Updated • 526
• 4
Sweaterdog/Andy-4-micro-base-LoRA-testbed
Updated
Sweaterdog/Smollest-reason-LoRA
Updated • 10
Sweaterdog/Andy-4-micro-0427-LoRA
Updated
Sweaterdog/Andy-4-micro-base-LoRA
Updated • 10
Sweaterdog/Smoller-reason2.1-ppo-LoRA
Sweaterdog/Smoller-reason2.1
Updated
Sweaterdog/Smol-reason2.1
3B • Updated • 376