GLM-5.2 arrives across GLM Coding tiers with a 1M-token context, two effort modes, Anthropic-compatible endpoints and no benchmarks at launch.
Export Controls Failed for PGP; Unlikely to Stop Anthropic’s Mythos
TechCrunch argues three decades of failed export controls on security software cast doubt on any attempt to block Anthropic’s Mythos model.
VibeThinker-3B Matches DeepSeek V3.2 and Kimi K2.5 on Verifiable Benchmarks
VibeThinker-3B is a 3B MIT-licensed dense reasoning model built on Qwen2.
Adds execute_write_sql tool to request approval before DB writes
datasette-agent 0.3a0 introduces execute_write_sql to prompt for user approval and apply DB permissions, plus chat CLI approval flags.
Export Controls Failed for PGP; Unlikely to Stop Anthropic’s Mythos
TechCrunch argues three decades of failed export controls on security software cast doubt on any attempt to block Anthropic’s Mythos model.
Google DeepMind Models Trained on Concept Art for Tribeca Film
At Tribeca, creators behind Dear Upstairs Neighbors used concept art to fine-tune custom builds of Google DeepMind’s Veo and Imagen, signaling a move away from prompt-only workflows.