Gawk · 2026-05-09

Gawk — 2026-05-09 · 8 tool incidents

Five verifiable things that moved in the AI ecosystem in the last 24h. Every number traces to a public source.

What moved

  • · Open-weight models hold 2 of the OpenRouter top 5 (vs 1 yesterday).
  • · claude-opus-4-6-thinking holds #1 on LMArena for the 18th consecutive snapshot.
  • · llamaindex downloads declined for the 14th consecutive snapshot.

Benchmark movers

#

No rank changes in the LMArena top 3

Why this matters · Public benchmarks are gameable, but rank shuffles still hint at where the frontier is genuinely moving.

Source: lmarena.ai

Tool Health

#

8 incidents in the last 24h (vs 2 yesterday)

Why this matters · Provider outages and degradations cause retry storms upstream. Tracking the 7-day shape catches flapping providers before they page you.

Tool health, 7 days ending 2026-05-09. Each row is one tool; each column is one UTC day. Green = operational, amber = degraded, red = outage, grey = no data.
  • Elevated errors on Claude Opus 4.1

    started 07:57 UTC · resolved 08:22 UTC · minor impact

    Anthropic status page

  • Claude Code IDE extension unable to load on Windows

    started 22:30 UTC · resolved 00:24 UTC · major impact

    Anthropic status page

  • Elevated errors on Claude Opus 4.7

    started 17:01 UTC · resolved 17:25 UTC · minor impact

    Anthropic status page

  • Elevated errors on Claude Sonnet 4.6

    started 15:03 UTC · resolved 15:11 UTC · minor impact

    Anthropic status page

  • Elevated errors across Claude Models

    started 09:49 UTC · resolved 11:40 UTC · major impact

    Anthropic status page

  • Elevated errors for Responses API

    started 00:07 UTC · resolved 00:08 UTC · major impact

    OpenAI status page

  • Degraded Performance with Codex Cloud Tasks

    started 15:07 UTC · resolved 16:17 UTC · minor impact

    OpenAI status page

  • Increased Error Rate for gpt-5.5 model in the API

    started 12:31 UTC · resolved 14:14 UTC · minor impact

    OpenAI status page

  • openai-api: Degraded → Operational

    Statuspage

  • codex: Degraded → Operational

    Statuspage

Source: status.anthropic.com, status.openai.com, status.openai-api.com, status.codex.com

Top HN stories

#

Top 5 on HN in the last 24h

Why this matters · What developers debate on Hacker News often previews which models, frameworks, and patterns they'll actually adopt next.

Source: news.ycombinator.com

SDK adoption

#

6 notable shifts across the six registries

Why this matters · Package install volume is where developers place real bets — distinct from where labs are shipping marketing.

Source: pypistats.org, npmjs.com, crates.io, hub.docker.com, formulae.brew.sh, marketplace.visualstudio.com

Notable AI Lab activity

#

4 labs moved in the last 24h

Why this matters · GitHub event volume on a lab's own repos is the cleanest publicly-verifiable proxy for engineering activity we have.

  • Microsoft Research

    −36 events (now 106) · Redmond, US

    Gawk · Labs

  • Google Research

    +77 events (now 85) · Mountain View, US

    Gawk · Labs

  • DeepSeek

    −5 events (now 69) · Hangzhou, CN

    Gawk · Labs

  • Google DeepMind

    +10 events (now 20) · London, GB

    Gawk · Labs

Source: gawk.dev

Model Usage

#

moonshotai/kimi-k2.6 holds #1 · biggest mover: deepseek/deepseek-v4-pro +16 to #8 · biggest drop: tencent/hy3-preview -37

Why this matters · OpenRouter rankings reflect API-first developer spend; direct customers like consumer ChatGPT are invisible by construction.

Source: openrouter.ai

Archived from . Every number traces to a public source.