Chandra OCR 2.6 is 2.5x faster than 2.4: median per-page time drops to 6.7s
VikParuchuri · x · 2026-10-28
Vik Paruchuri shared speed details for Chandra OCR 2.6: 2.5x faster than version 2.4, with median per-page time dropping from 17.0s to 6.7s and p90 from 33.3s to 14.1s, measured on the same 1,187 OmniParseBench pages. Try it on the Datalab website; free credits available on signup.
Related event: Datalab's Chandra OCR 2.6 beats frontier models at 5-15x lower price(3 posts)→
More from Models
- FrontierCode 2.0: Claude Opus 5.5 tops private coding benchmark at 46.4%, Kimi and DeepSeek close behind — bookwormengr · 2026-10-28
- Steering Vector Recipe Turns Qwen 27B Into "Evil Qwen" — basedjensen · 2026-10-28
- MoveBench debuts: a rubric-based benchmark for LLM sports and health coaching — Yitong Li · 2026-10-28
- Debate: Claude isn't 'lobotomized' — inference stack updates may shift old-model outputs — JasonBotterill · 2026-10-28
- AI labs near IPO, yet LLMs still produce 'astonishingly terrible prose' by default — GabGarrett · 2026-10-28
- Rebuttal: Kimi 3.1 Nears Fable on Frontier Code 2, Contradicting 'China Models Crushed' Claim — bookwormengr · 2026-10-28