Hacker News · points history

OCR It – pull text out of un-copyable documents for your LLM

SteadyVane has taken 31 readings of OCR It – pull text out of un-copyable documents for your LLM since Aug 24, 2026: 2 points at the first reading, 139 at the latest (+137). Its biggest one-day gain was +114 on Aug 24, 2026.

open the story · Hacker News thread

first seen points

139

peak points

139

latest points

observations

first seen Aug 24, 2026 · last seen Aug 29, 2026 · published Aug 24, 2026

Day by day

Day (UTC)	points	Change
Aug 29, 2026	139	0
Aug 28, 2026	139	0
Aug 27, 2026	139	0
Aug 26, 2026	139	+1
Aug 25, 2026	138	+22
Aug 24, 2026	116	+114

Each day's last reading. Change: against the day before (after days with no reading: since the last day that has one); the first day shows the change within that day.

On the tracker; not yet featured in a brief.

More rising on Hacker News

Anthropic appears to be A/B testing reduced effort levels in Claude Code+16452 → 216 points
Check if a file was made with Claude+15927 → 186 points
How well do agents use test/verification techniques?+14249 → 191 points
Models Are Getting Dumber on Purpose+139196 → 335 points
Claude Cowork and chat are now one Claude+132101 → 233 points
Show HN: ThoughtDAG – An editable context graph for LLM conversations+1324 → 136 points
OpenAI expands ChatGPT ads with Sponsored Agents+12136 → 157 points
OpenAI's GPT-6 Astra on ARC-AGI-3+117121 → 238 points

Everything rising on Hacker News →