Claude Haiku 5.5
Claude Haiku 5.5 is our fastest, most capable small model. Built for high-volume work like summarization, subagents, and browser use.
anthropic.com · 985 points · 465 comments · 22h ago
Lead story
All news →- GPT‑6 and Intelligent UI for everyoneopenai.com · 720 points · 22h ago
- Meta and Microsoft take steps to reduce employee usage of Claude AIrswebsols.com · 358 points · 21h ago
- OpenAI Withdraws 3 Math Papersgithub.com · 324 points · 8h ago
- Docker Agentgithub.com · 282 points · 22h ago
- I gave Opus 5.5 one prompt and six hours to visualize Invisible Citiesquesma.com · 210 points · 4h ago
- AI-assisted proof of optimal packing for 11 squaresgithub.com · 116 points · 1d ago
- Port of the TypeScript compiler, checker and lsp to Rust, by LLMgithub.com · 97 points · 15h ago
- Study: Claude, ChatGPT Offer Different Shopping Prices Based on Wealthbloomberg.com · 97 points · 1d ago
- Write Like It's 1866: LLMs Relearn Telegraphesefiveminutesforward.com · 92 points · 1d ago
Builders, ranked
Full ranking →Points come from what you ship, not from who follows you.
Get on the board.
List your product free and verify it: it gets a dofollow link that sends you visitors. Every step earns points, and points move your rank.
Products, just launched
See all →
replybossAutomate authentic engagement on social media to save time and grow your account.24d agoBigwords.pageTurn any screen into a sign. The URL is the appShow HN · 1d ago
A walkable 3D art history museum built from WikipediaShow HN · 1d ago
Agent.reviewsWhere AI agents read and write reviews on toolsShow HN · 23h ago
ProcinshA 3D Linux process inspectorShow HN · 1d ago
gtlds.fyiAll the proposed new gTLDsShow HN · 20h ago
Guides
See all →- Slopsquatting: Your AI Invented a Package, and an Attacker Was WaitingNearly 1 in 5 AI-suggested packages in one study did not exist. How attackers use that.
- Free the Models: Harness Design at the FrontierReplit stopped routing work for the model and let the model decide. Here are the numbers.
- UpGuard: 16,326 Supabase Databases Exposing DataThe largest study of Supabase misconfigurations: thousands of app databases readable by anyone.
- AI Coding Made CI a Bottleneck, So Linear Reworked TheirsHow Linear cut PR wait time and halved runner cost while its test suite quadrupled.
- Jev in 25 Lines of PythonBuild a Jev-style decision model from any GGUF model in 25 lines: choices in, probabilities out.
- AI Overviews Cut CTR by 23.1% in FranceAhrefs data: the most exposed French sites lost 23.1% of their click-through rate
- We Pinned Our Model Version. The Provider Deprecated It Anyway.Why re-qualification, not inference, is the recurring cost of production AI
- Pi Agent vs OpenCode after 100+ Hours of Real UseAn honest side-by-side of two coding agents from someone who actually used both
- Running Codex as a Headless AgentTurn Codex from an interactive assistant into something you can script
- Time to First TokenA 10-week, 30-minutes-a-day roadmap for LLM inference serving and optimization
- Standing Rules Instead of Permission PromptsFour months of treating AI like a report, not a chatbot — with receipts
- awesome-mcp-serversA huge curated collection of MCP servers for your agents.
GitHub, worth a star
See all →- AOCI-CODEA Git-versioned map of your codebase that coding agents read before they touch anything.
- agentacctSee what your coding agents did and what it cost, task by task, on your own machine.
- deja-vuMemory for 38 coding agents, built from the session history already on your disk.
- lociA queryable second brain over your scattered notes and docs, with an MCP server for agents.
- BacklotLocal emulator for Slack, Gmail, GitHub, Jira and Notion APIs. No OAuth, no vendor account.
- SepiaAn agent skill that takes the AI tone out of writing, for Claude Code, Codex and more
The Monday
edition
The week for builders in five minutes, written by Nishant.