← 최신 프로덕트

Zenith

sota harness for normal models to beat Fable on FrontierSWE. Zenith is our long-running agent harness. On Frontier SWE — seventeen of the hardest public long-horizon software-engineering tasks we know of — Zenith took GPT-5.5 from 5th place to 1st, ahead of Claude Fable, by building the right harness around the model instead of swapping in a bigger one.

ii.inc 방문외부 사이트

반응 (1)

비슷한 프로덕트

설명이 비슷한 순

  1. Mingbird

    github.com

    an agent harness that makes a 2B model finish real tasks

    Hacker News5점

  2. HarnessRouter

    github.com

    Unified interface for agent harnesses

    Hacker News10점

  3. Unified access to top AI models

    labophase.com

    supporting GPT4, Claude and more

    Hacker News19점

  4. Raven

    github.com

    The harness of harnesses, built for RSI

    Hacker News55점

  5. I built an open source multi-agent harness in Go

    github.com

    Open source multi-agent harness for orchestrating AI workers. Supports Claude Code, OpenAI Codex, Cursor Agent, and local models.

    Hacker News6점