S1-Omni Unifies Scientific Multimodal Reasoning
Jiahao Zhao · hf · 2026-07-20
S1-Omni: A Unified Scientific Multimodal Reasoning Model
S1-Omni is a unified multimodal reasoning model designed for AI for Science, aiming to integrate scientific understanding, prediction, and generation within a single model. It maps natural language instructions alongside CIF, SMILES, protein sequences, spectra, and scientific images into a shared representation space, training the model with knowledge of both the natural world and scientific laws.
The authors state that the S1-Omni-Corpus covers 200 scientific tasks, contains millions of reasoning samples, and is evaluated on 60+ scientific benchmarks. Reportedly, it surpasses GPT-5.5 and Gemini-3.1-Pro on most benchmarks and matches or exceeds domain-specific models on several tasks.
Supported tasks include:
- Property prediction
- Spectrum-to-molecule generation
- Protein site and structure prediction
- Scientific image generation and editing
Related event: S1-Omni: A Unified Multimodal Model for Scientific Reasoning(2 posts)→
More from Multimodal
- One prompt to a 2-3 min narrated short: hands-on test of Antigravity video generation — AI_Andrew · 2026-10-26
- FLUX.3 Klein step-distilled to 4-8 steps with CFG baked in — Fit-Structure-6489 · 2026-10-26
- Full animation short made with Google Gemini tools, 10,564 frames — AI_Andrew · 2026-10-26
- ostris pre-training "Primal VAE": 1-channel latents match modern 8x VAEs via patching — ostrisai · 2026-10-26
- Community checkpoint 'Narrative Drift' brings art-directed visuals to Krea 2 Turbo — MoonbearAIArt · 2026-10-26
- Same prompt, split results: Claude Opus 5.5 vs GPT-6 Astra on a 3D C-section explainer — Sourcecode12 · 2026-10-26