Blog
Learn moreMore Realistic Images, More Stable Editing Across 100 Rounds | Introducing Seedream 5.5
Pixel-level preservation of unedited regions, with major improvements in realism and small-text rendering.
SeedRealtime Audio-Visual Full-Duplex LLM Released: Toward Omni-Modal Natural Interaction
As a native audio-visual full-duplex LLM, SeedRealtime can jointly understand audio, visual, and temporal information, accurately identify the interaction target and user intent, delivering a brand-new "watch, listen, and speak" experience.
Aug 5, 2026Learn more
One-take Creation, Flexible Referencing: Introducing Seedance 2.5
30-second long-form storytelling with fully upgraded multimodal referencing and editing capabilities.
Jul 31, 2026Learn more
From Speech to Audio Creation | Introducing the Seed Audio 1.0 Audio Creation Model
A unified framework for joint modeling of voice, sound effects, ambient sound, and other audio elements.
Jul 20, 2026Learn more
EdgeBench: Measuring Real-World Environment Learning and Discovering a New Scaling Law
Agents exhibit a clear log-sigmoid pattern in environment learning.
Jul 7, 2026Learn more
Publications
View allOct 25, 2026Recycler: Cost-Free Latent Reasoning via Hidden State Recycling
Sep 28, 2026Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression
Jul 6, 2026EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments
Jun 30, 2026Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
May 29, 2026Task-Focused Memorization for Multimodal Agents
May 16, 2026Charon: A Unified and Fine-Grained Simulator for Large-Scale LLM Training and Inference
Apr 26, 2026MegaScale-Omni: A Hyper-Scale, Workload-Resilient System for MultiModal LLM Training in Production
Apr 22, 2026Seed3D 2.0: Advancing High-Fidelity Simulation-Ready 3D Content Generation
Apr 15, 2026Seedance 2.0: Advancing Video Generation for World Complexity
Apr 11, 2026Protenix-v2: Broadening the Reach of Structure Prediction and Biomolecular Design