News archive
All posts and signals, newest first.
SearchClaude Code users, change this setting if you use subagents
Claude has prompt caching with a 1-hour TTL, but that’s only for the main agent. Subagents get just 5 minutes by default. If you use long-running…
Microsoft-Decision-1: API availability
Step 5 Preview Free on OpenCode Zen
Get your banked reset and sign for it
Claude Sonnet is now 20% cheaper on many tasks thanks to cheaper cache
GPT-6 is rolling out in web, including to Plus users!
Claude Haiku 5.5 is out!
Absolutely destroys GPT-Luna across all benchmarks. The model is extremely cheap, especially under 100k context. Input $0.10 / MTok up to 100,000…
Anthropic announced Claude Haiku 5.5
Claude Haiku 5.5 is out
Claude Haiku 5.5 appeared in the new Claude Code binary
Tibo reset Codex limits
The features were pretty weak. Even the new math proofs release couldn’t save it. Series tied 1-1.
ChatGPT Intelligent UI: Product update
Claude Haiku 5.5: API availability
Day 2 of the OpenAI sprint: negative overall
2.1 is a decent feature. Auto-review for risky actions sounds useful, especially for sub-agents and long-running tasks. But I’m always on bypass…
The French at Mistral just dropped Large 4
Looks like Europe’s biggest hope in frontier AI is still alive. 1 trillion parameters, with 49B active at a time. In some benchmarks, it already…
Decisions API: Catalogue listing
EmbeddingGemma 2: Open weights
Nano Banana 2.1: API availability
Mistral Large 4: Public preview
Eleven v4 Turbo: API availability
Day 1 of OpenAI’s sprint: overall positive
Going from ~30 TPS to ~50 TPS is already a meaningful improvement and basically puts it at Fable-level speed. That said, I think the bar is higher…
The OpenAI team has entered its redemption arc
From now on, every day brings either a genuinely useful feature or a full reset. Let the comeback begin.













