1d agoLisanBench analysis finds Claude Opus 4.8 at 1-million-token context approaches GPT-5.5's 256K scores on GraphWalks— GraphWalks BFS scores jumped to 68.1% from v4.7's 40.3%.——0——Original postOPLA#980Lisan al Gaib|@SCALING01Long-Context capabilities are steadily increasing from Opus 4.6 to Opus 4.8 Opus 4.8@1 million now almost as good as GPT-5.5's 256K score10:15 AM · May 28, 2026 View on XPOSTAV#1245Alex Volkov|@ALTRYNEOpus 4.8 should be MUCH better at long context, the jump is quite high on GraphWalks especially in the 1M subset!9:53 PM · May 28, 2026 · 1.2K Views
POSTAV#1245Alex Volkov|@ALTRYNEOpus 4.8 should be MUCH better at long context, the jump is quite high on GraphWalks especially in the 1M subset!9:53 PM · May 28, 2026 · 1.2K Views
