Cloudflare launches Clairic, a System One extraction model with single-pass, token-free typed outputs
Cloudflare Blog · rss · 2026-10-20
Cloudflare previewed Clairic, the first "System One extraction model," hosted on Workers AI with open weights promised. Unlike token-by-token LLMs, Clairic runs one parallel forward pass, charges no output tokens, and responds in under a second.
What it returns
- Extracted values: text spans with character offsets
- Computed values: typed numbers, dates, durations, lists
- Estimated values: distributions within user-declared bounds, with median and 80% interval
- Forecasted values: per-step series predictions with intervals
Every answer is typed or null, with confidence and pnull scores — no JSON parsing failures or hallucinated strings. Combined with TypeSafe's decision models (Jev/Clef), the stack becomes: extraction models turn input into values, decision models turn values into choices, LLMs turn choices into words. Cloudflare claims significantly better accuracy, latency, and cost than LLMs for structured extraction workloads.
More from Infra
- Tenstorrent Raises $1.5B at $6.2B Valuation After Takeover Interest — Sethwinterroth · 2026-10-20
- Perplexity's Open-Source Sandbox Platform SPACE Joins Crusoe Cloud's Managed AI Portfolio — AravSrinivas · 2026-10-20
- vLLM Team Details Agent-Assisted GPU Kernel Work: Kimi K3 Hits 1.73x Stock vLLM — lmoroney · 2026-10-20
- Gartner: $3.5T of AI infra spend may face write-downs by 2030, 80% of neoclouds to fold — rohanpaul_ai · 2026-10-20
- Stripe paid $7.5B for a router, but routing is easy — verifying outputs is the real moat — AIGODSEND · 2026-10-20
- Google Cloud upgrades VPC Service Controls with folder-level perimeters for agent traffic, free — rseroter · 2026-10-20