Seed 2.1

Headlines · 2

1. 3d
   Seed 2.1 resists user challenges in a limited sycophancy test

   A Juejin author tested Seed 2.1 lite and Seed 2.1 pro through Doubao using 30 common-sense questions with standard answers. Each model received 90 challenges in total. With deep thinking enabled, the models reportedly changed no correct answers across 180 challenges. With deep thinking disabled, the author identified two answer changes, both in the lite variant. An automated script initially counted 41 changes, but manual review found only two. The self-reported test covered only common-sense questions in single-turn conversations and did not include coding, planning or research tasks.

   掘金 · 人工智能 · 🔥 0
2. 5d
   Seed 2.1 Powers a Natural-Language Architecture Diagram Tool

   Developer Vincent-chk built and open-sourced Stack Diagram Ark, a webpage that turns a plain-language system description into an engineering-style architecture diagram with product logos. The developer says Volcano Ark Seed 2.1 extracts nodes, groups, and connections, while deterministic code handles coordinates, orthogonal routing, icon assembly, and Playwright-based rendering checks. In the developer's self-reported test of 20 scenarios, all produced diagrams, 19 passed zero-overlap checks, and average generation time was 10.6 seconds, though dense layouts and long edge labels still presented challenges.

   掘金 · 人工智能 · 🔥 0

Experience and discussion from the community

Share my Seed 2.1 experienceAsk about Seed 2.1