Firecrawl Raises $75M Series B: Alexandria for AI Agents Firecrawl closed a $75M Series B led by Smash Capital to launch Alexandria—a unified knowledge library for AI agents. What it is: Alexandria connects official data providers, custom connectors, and Firecrawl's own indexes (research, developer, government) so AI agents find, understand, and pull from any source in one API call. The metrics: • 1.5M users already building with Firecrawl • 21% higher answer quality vs. built-in web tools (blind testing, 845 tasks) • Pays data providers (Wikimedia Enterprise first) • Covers tens of million

NEW MODEL CLASS: Jev (TypeSafe AI) ChatGPT co-inventor Diogo Almeida just shipped a model that refuses to generate text. Why? Because LLMs are built for chat. Jev is built for decisions. Key metrics: - 40-200x faster than GPT-6 on automation (70-500ms) - 444x cheaper per decision - Zero hallucinations, zero type errors - Parallel sampling vs sequential tokens - $40M Series A from DCVC How it happened: 2 years stealth → identified gap in automation → built new architecture + RLCD training → proved on workflows (Doom, Wikipedia racing) → raised $40M → launched Sept 15. Why it matters: Jev sl

Tokens Too Cheap to Meter: AI Inference Cost Collapse AI inference costs dropped 2.5 orders of magnitude in 2026 alone. Models are 100x more cost-efficient per task. GPU efficiency doubles every 2 years. Specialized models like Jev now cost $42 per billion tokens—cheaper than electricity. By 2027, running AI will cost less than a single tool call. This inverts the game: AI becomes infrastructure, not product. Incumbents with switching costs (healthcare, finance) win. Startups must compete on quality and UX, not raw capability. Your moat isn't inference. It's product-market fit. https://jyn