technique

context compaction

techniqueactiveprovisionalcontext-compaction-68f498bd·1 events·first seen 15d ago

Aliases: context compaction

Co-occurring entities

GPT-5.2 Claude Opus 4.6 adaptive thinking BrowseComp Humanity's Last Exam Cowork Claude Code OpenAI GDPval-AA Anthropic Terminal-Bench

More like this (12)

thought compression COMPACT-VA context-mode Context-Driven Incremental Compression BrowseComp-Plus gradient compression context governance End-to-End Context Compression at Scale BrowseComp Agent JIT Compilation CompVis Planning-aligned Token Compression for Long-Context Autonomous Driving

Recent events (1)

9Anthropic News·15d ago·source ↗

Claude Opus 4.6 Released with 1M Token Context, Agentic Coding Advances, and State-of-the-Art Benchmarks

Anthropic has released Claude Opus 4.6, its most capable model to date, featuring a 1M token context window in beta, improved agentic coding and planning capabilities, and adaptive thinking with developer-controlled effort levels. The model claims top scores on Terminal-Bench 2.0, Humanity's Last Exam, GDPval-AA, and BrowseComp, outperforming OpenAI's GPT-5.2 by 144 Elo points on GDPval-AA. New product features include agent teams in Claude Code, context compaction for long-running tasks, and Claude in PowerPoint (research preview). Pricing remains unchanged at $5/$25 per million input/output tokens.

Long Context Evolution Frontier Model Releases GPT-5.2 Claude Opus 4.6 adaptive thinking +13 more