Seed1.8 Model Card: Towards Generalized Real-World Agency
Research paper by Bytedance Seed. arXiv record dated 2026-04-17. Research area: Artificial intelligence.
Resource information
- Authors
- Bytedance Seed
- Research area
- Artificial intelligence
- Application domain
- General AI research
- AI methods
- Large language models
- Research topics
- cs.AI
- arXiv record date
- 2026-04-17
- Publication
- arXiv
- Upstream last modified
- Apr 20, 2026
- Last record update
- Nov 4, 2026
Abstract
We present Seed1.8, a foundation model aimed at generalized real-world agency: going beyond single-turn prediction to multi-turn interaction, tool use, and multi-step execution. Seed1.8 keeps strong LLM and vision-language performance while supporting a unified agentic interface-search, code generation and execution, and GUI interaction. For deployment, it offers latency- and cost-aware inference, including configurable thinking modes and optimized visual encoding for images and video. We report evaluations on standard benchmarks and application-aligned workflows spanning foundational skills, multimodal understanding, and agentic behavior. Seed1.8 is released to support further research and development on interactive, real-world use cases.