The Open Weights
LatestModelsLeaderboardsCompanies
Subscribe
The Open Weights

The daily record of open-source AI. New model releases, leaderboards, and what's coming next — written for people who ship.

Refreshed every 12 hours

Discover

  • Latest releases
  • New today
  • Trending models

Browse

  • All models
  • Companies
  • Categories
  • Leaderboards

About

  • About
  • Editorial policy
  • Hardware estimates
  • RSS feed
  • llms.txt
  • Newsletter

© 2026 The Open Weights. An independent publication.

PrivacyTermsSMSAggregated by Claude · curated by humans.
LatestSperidlabsIDOL
SperidlabsText → 3D

IDOL brings image-to-3D Gaussian splatting to Apache

Sperid Labs releases IDOL, an open model that turns images into 3D Gaussian-splat representations under a permissive license.

Oct 28, 2026
NotableApache 2.0
IDOL

Sperid Labs has published IDOL, a model that reconstructs 3D scenes from imagery using Gaussian splatting, now available on Hugging Face under the permissive Apache-2.0 license.

Gaussian splatting has become one of the most active areas in 3D graphics and vision, offering a way to represent scenes as collections of 3D Gaussians that can be rendered quickly and with high fidelity. IDOL applies that approach to the image-to-3D problem, taking visual input and producing a splat-based 3D representation.

Why it matters

Open releases in the 3D space remain comparatively rare next to the flood of text and image models, so a permissively licensed entry carries weight for researchers and builders working on reconstruction and content creation.

  • Primary modality: text-to-3D, with an image-to-3D workflow
  • License: Apache-2.0, allowing commercial use and modification
  • Hosted on Hugging Face for direct access

Sperid Labs has not published detailed benchmarks or parameter counts alongside this initial release, so independent evaluation will be the real test. For teams experimenting with 3D asset generation, IDOL offers another open foundation to build and compare against.

Sources

  • speridlabs/idol

    Hugging Face

    Visit
OlderinclusionAI releases Ling-3.1-flash, an MIT MoE modelinclusionAI · Text / LLM · last weekNewerServiceNow Debuts Apriel Voice 27B, an Open Speech ModelServiceNow · Text → Speech · last week

Get the model

Hugging Face

Specs

LicenseAPACHE-2.0
Downloads14.6K
Likes618

Modalities

Text → 3D

The Weekly Weights

Every open release that mattered, one email a week.

0 comments

No comments yet. Be the first to weigh in.

More in Text → 3D

All Text → 3D →
HY-WorldPlay
Tencent/Image → Video

Tencent's HY-WorldPlay Creates 3D Scenes from One Image

The new model from Tencent's Hunyuan team generates dynamic video and reconstructs 3D environments using a single static picture.

Dec 12, 2025
HunyuanWorld-Voyager
Tencent/Image → Video

Tencent's Voyager Model Turns Images into 3D Worlds

The new model from Tencent AI Lab generates temporally and spatially consistent video sequences from a single image, enabling virtual exploration of static scenes.

Aug 27, 2025