Live telemetry · daily brief
GAWK
by nativerse
● ARCHIVE
2026-09-14
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 9th consecutive snapshot.
- ■openai downloads declined for the 3rd 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. pypistats.org, crates.io, hub.docker.com · view all →
+366.9k
Docker Hub: ollama/ollama
+366.9k all-time pulls day-over-day
+8.8k
VS Code Marketplace: saoudrizwan.claude-dev
+8.8k cumulative installs day-over-day
Model Usageopenai/gpt-5.6-luna holds #1 · biggest mover: meta/muse-spark-1.3-contributor +13 to #12 · biggest drop: google/gemini-3.7-flash -13
OpenRouter rankings reflect API-first developer spend; direct customers like consumer ChatGPT are invisible by construction. openrouter.ai · view all →
+13
meta/muse-spark-1.3-contributor
was #25 on 2026-09-07
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.
-13
google/gemini-3.7-flash
was #12 on 2026-09-07
—
#1 openai/gpt-5.6-luna
#1 openai/gpt-5.6-luna
—
#2 tencent/hy4-preview
#2 tencent/hy4-preview
—
#3 z-ai/glm-5.3-flash
#3 z-ai/glm-5.3-flash
Tool Health
2 incidents in the last 24h (vs 1 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 error rates for Codex and ChatGPT Work
started 02:46 UTC · resolved 03:57 UTC
Elevated errors for ChatGPT users in Europe
started 16:56 UTC · resolved 17:47 UTC · minor impact
Benchmark movers
3 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
-2 Elo (now 1508) · anthropic
claude-opus-5-max
up 1 to #2 · +1 Elo (now 1505) · anthropic
claude-opus-5-high
down 1 to #3 · 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.
Claude Fable 5.1 Solves the Cyphral Distich, a 370-year-old cipher
179 points · 47 comments
Astra and Fable still hack on simple variants of alignment evals from 2025
114 points · 31 comments
David Sacks: OpenAI and Anthropic Don't Need Regulations to Pace Frontier Models
105 points · 70 comments
Houthis Used Claude Code to Develop Missile Guidance Software: Anthropic
62 points · 59 comments
Due to concerns about malicious applications, GPT2 will not be released (2019)
59 points · 28 comments
Notable AI Lab activity
10 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 195) · San Francisco, US
OpenAI
+22 events (now 138) · San Francisco, US
Anthropic
+5 events (now 123) · San Francisco, US
Ollama
+6 events (now 100) · San Francisco, US
Hugging Face
+6 events (now 93) · New York, US
Microsoft Research
+31 events (now 66) · Redmond, US
DeepSeek
+9 events (now 37) · Hangzhou, CN
INRIA
+15 events (now 29) · Paris, FR
New mover: Google Research
11 events · Mountain View, US
Dropped off: Weights & Biases
was 9 events · San Francisco, US
Get this board in your inbox — one email, every morning (UTC).
Subscribe free