← Trending

Tangerine Index lists products launched since ChatGPT and people's reactions to them.

Cognition's SWE-2

cognition.com

Cognition's coding model, 64% cheaper than Fable 5.1

Visit cognition.comexternal site

In the makers' words

Today we’re introducing SWE-2, our most advanced coding model yet. SWE-2 delivers highly competitive agentic coding performance across multiple effort…. SWE-2 is Cognition's new coding model, post-trained from Kimi K3 with RL that optimizes for cost and capability at the same time.

Reactions (1)

Similar products

by description similarity

  1. Claude Fable 5.1

    Claude’s most advanced models for coding and knowledge work

    anthropic.comlatest reaction

    Hacker News
    No reaction
    Product Hunt
    146 votes
    GitHub
    No reaction

  2. Edgee Codex Compressor V2

    Use Codex at 35.6% lower costs

    edgee.ai4 reactionslatest reaction

    Hacker News
    No reaction
    Product Hunt
    212 votes
    GitHub
    No reaction

  3. Komi-learn

    continuous memory and self-improvement for coding agents

    github.comOpen source2 reactionslatest reaction

    Hacker News
    27 pts
    Product Hunt
    No reaction
    GitHub
    80 stars

  4. Kimi K2.7 Code

    Kimi’s most capable coding model yet

    huggingface.colatest reaction

    Hacker News
    No reaction
    Product Hunt
    336 votes
    GitHub
    No reaction

  5. A plain-text cognitive architecture for Claude Code

    One memory, not one per tool. Cog gives AI agents a shared, structured memory layer using plain-text files — portable across Claude Code, Codex, Cursor, Gemini CLI, Windsurf, Copilot, and more.

    lab.puga.com.brlatest reaction

    Hacker News
    157 pts
    Product Hunt
    No reaction
    GitHub
    No reaction