Live telemetry · daily brief
gawk.dev
by nativerse
● ARCHIVE
2026-09-22
2 tool incidents
What verifiably moved in the AI ecosystem in the last 24h. Every number traces to a public source.
What moved
- ■claude-fable-5.1-max holds #1 on LMArena for the 17th consecutive snapshot.
- ■@anthropic-ai/sdk downloads declined for the 4th consecutive snapshot.
- ■openai downloads declined for the 4th consecutive snapshot.
SDK adoption6 notable shifts across the six registries
Package install volume is where developers place real bets — distinct from where labs are shipping marketing. pypistats.org, npmjs.com, crates.io · view all →
+383.5k
Docker Hub: ollama/ollama
+383.5k all-time pulls day-over-day
+15.7k
VS Code Marketplace: saoudrizwan.claude-dev
+15.7k cumulative installs day-over-day
Model Usagez-ai/glm-5.3-flash holds #1 · biggest mover: openai/gpt-6-astra +13 to #14 · biggest drop: anthropic/claude-sonnet-4.6 -17
OpenRouter rankings reflect API-first developer spend; direct customers like consumer ChatGPT are invisible by construction. openrouter.ai · view all →
+13
openai/gpt-6-astra
was #27 on 2026-09-15
OpenRouter request volume reflects developer API spending, not end-user adoption. Biased toward API-first workflows that route through OpenRouter — direct OpenAI / Anthropic / Google customers who never use OpenRouter are invisible.
-17
anthropic/claude-sonnet-4.6
was #31 on 2026-09-15
—
#1 z-ai/glm-5.3-flash
#1 z-ai/glm-5.3-flash
—
#2 deepseek/deepseek-v4.1-flash
#2 deepseek/deepseek-v4.1-flash
—
#3 tencent/hy4-preview
#3 tencent/hy4-preview
Tool Health
2 incidents in the last 24h (vs 0 yesterday)
Why this matters · Provider outages and degradations cause retry storms upstream. Tracking the 7-day shape catches flapping providers before they page you.
Elevated errors for multiple models
started 00:57 UTC · resolved 02:35 UTC · major impact
Increased error rate for Plus and Pro users.
started 09:58 UTC · resolved 10:37 UTC · minor impact
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.
#1 claude-fable-5.1-max
anthropic · 1508 Elo
#2 claude-opus-5-max
anthropic · 1505 Elo
#3 claude-opus-5-high
anthropic · 1505 Elo
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.
Amazon blocks Meta’s new Muse AI agent from shopping on amazon.com
77 points · 67 comments
Claude Status – Elevated errors for multiple models
75 points · 60 comments
Transformers Explained Visually
64 points · 7 comments
AI coding has made CI a bottleneck, so we reworked ours to keep up
53 points · 28 comments
JetBrains Air: A System of Products for Agentic Software Development
46 points · 38 comments
Notable AI Lab activity
11 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.
Anthropic
+14 events (now 142) · San Francisco, US
Ollama
+9 events (now 98) · San Francisco, US
OpenAI
−17 events (now 89) · San Francisco, US
Hugging Face
+5 events (now 88) · New York, US
New mover: Weights & Biases
88 events · San Francisco, US
Microsoft Research
−6 events (now 42) · Redmond, US
DeepSeek
+12 events (now 41) · Hangzhou, CN
Alibaba Qwen
+12 events (now 27) · Hangzhou, CN
New mover: Intel AI
16 events · Santa Clara, US
Dropped off: Meta FAIR
was 13 events · Menlo Park, US
Dropped off: Google Research
was 12 events · Mountain View, US
Get this board in your inbox — one email, every morning (UTC).
Subscribe free