Geo News
Community curated by people like you
LatestAICryptoHealthWorld AffairsUS Politics

Model evaluation methodology stories

Oct 2, 2026

Cloudflare releases Clef decision models for AI agent classification

Cloudflare released Clef and Clef-flash, open-weight decision models that return typed probabilities instead of text. Clef-flash delivers classifications in about 39 milliseconds, making it more than ten times faster than competing models.

Oct 2, 2026·2 sources
00
Aug 15, 2026

New benchmark shows AI models struggle with visual perception, none reach 60% accuracy

Moonshot AI's PerceptionBench reveals that leading multimodal AI models, including GPT-5.6 Sol, perform poorly at basic visual perception tasks when separated from logical reasoning, with no frontier model achieving 60% accuracy. The benchmark demonstrates that many errors attributed to reasoning actually occur during the image-reading stage.

Aug 15, 2026·1 source
00
Jul 24, 2026

Snorkel AI Highlights First Wave of Open Benchmarks Grants Projects

Snorkel AI has revealed the initial projects receiving support from its Open Benchmarks Grants, a $3 million initiative dedicated to funding open-source datasets, benchmarks, and evaluation tools.

Jul 24, 2026·1 source
00

Top claims

  • ▪Cloudflare used its own variant of Reinforcement Learning for Calibrated Decisions to train Clef, a method that TypeSafe AI previously used for Jev
  • ▪Low latency is more important than deep reasoning capability for operational AI agents
  • ▪Cloudflare's Clef supports a 65,536-token context window and image inputs, whereas TypeSafe AI's Jev is limited to text inputs and a 32,000-token context window

Subtopics

AI research & benchmarks2Open-source AI2AI agents1AI inference scaling1AI safety benchmarks1AI startups1AI tools & products1Domain specific AI benchmarks1Multimodal models1Open model deployment tools1

Related timelines

Congress

110 stories

Crypto hacks

100 stories

Iran War

209 stories

Payments

118 stories

Russia-Ukraine war

136 stories