Live telemetry · daily brief
GAWK
by nativerse
● ARCHIVE
2026-09-12
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 7th consecutive snapshot.
SDK adoption5 notable shifts across the six registries
Package install volume is where developers place real bets — distinct from where labs are shipping marketing. npmjs.com, crates.io, hub.docker.com · view all →
+357.0k
Docker Hub: ollama/ollama
+357.0k all-time pulls day-over-day
+18.8k
VS Code Marketplace: saoudrizwan.claude-dev
+18.8k cumulative installs day-over-day
Agent frameworks1 agent framework moved >10% in the last week
Agent frameworks rise and fall on weekly download velocity and open-issue trajectory, not GitHub stars. pypistats.org · view all →
-14%
LangGraph · -14%
13M weekly downloads · 42k stars
Model Usagetencent/hy4-preview holds #1 · biggest mover: meta/muse-spark-1.3-contributor +24 to #13 · biggest drop: openai/gpt-oss-20b -17
OpenRouter rankings reflect API-first developer spend; direct customers like consumer ChatGPT are invisible by construction. openrouter.ai · view all →
+24
meta/muse-spark-1.3-contributor
was #37 on 2026-09-05
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
openai/gpt-oss-20b
was #32 on 2026-09-05
—
#1 tencent/hy4-preview
#1 tencent/hy4-preview
—
#2 openai/gpt-5.6-luna
#2 openai/gpt-5.6-luna
—
#3 z-ai/glm-5.3-flash
#3 z-ai/glm-5.3-flash
Tool Health
2 incidents in the last 24h (vs 5 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 Claude Mythos 5.1 and Claude Fable 5.1
started 14:19 UTC · resolved 14:37 UTC · major impact
1% of ChatGPT Work (mobile/web) turns are failing for existing threads
started 16:22 UTC · resolved 01:57 UTC · minor impact
openai-api: Degraded → Operational
Benchmark movers
2 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.
claude-fable-5.1-max
-4 Elo (now 1510) · anthropic
claude-opus-5-max
-1 Elo (now 1504) · anthropic
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.
OpenAI agents carried out an undisclosed attack on RubyGems
422 points · 244 comments
Top mathematicians are outraged by OpenAI's methods
88 points · 13 comments
HuggingFace: Security.txt
80 points · 7 comments
A misalignment of AI in mathematics
53 points · 149 comments
A Misalignment of AI in Mathematics
51 points · 21 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.
LangChain
+11 events (now 192) · San Francisco, US
Anthropic
−6 events (now 119) · San Francisco, US
OpenAI
−8 events (now 116) · San Francisco, US
Hugging Face
−25 events (now 103) · New York, US
Microsoft Research
+34 events (now 94) · Redmond, US
Ollama
−12 events (now 79) · San Francisco, US
Weights & Biases
−12 events (now 60) · San Francisco, US
DeepSeek
−10 events (now 21) · Hangzhou, CN
New mover: Google Research
10 events · Mountain View, US
Meta FAIR
−7 events (now 10) · Menlo Park, US
Dropped off: Intel AI
was 24 events · Santa Clara, US
Get this board in your inbox — one email, every morning (UTC).
Subscribe free