# Octopus paper shares one KV cache across sub-agents and lifts BrowseComp-Plus accuracy to 49.6% on a 4-billion-parameter model

The same week, a second paper made reproducible inference cheaper under batching, and Anthropic cut the cost of Claude Code's background subagent status updates by about 99% after the tool had re-read whole transcripts every 30 seconds.

- Published: 2026-11-06T10:11:19.661Z
- Canonical: https://polylog.news/ai/2026-11-06/octopus-paper-shares-one-kv-cache-across-sub-agents-and-lift
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv cs.LG](https://arxiv.org/abs/2611.06956), [arXiv cs.AI](https://arxiv.org/abs/2611.06977), [Polylog editors](https://polylog.news), [daily.dev](https://daily.dev/posts/claude-code-cuts-background-subagent-token-cost-by-99-in-v2-1-316-urwybiilg)

Multi-agent systems pay for context more than once. Octopus, a paper by Tianming Sha submitted November 3, identifies three costs in current frameworks: they copy prefix KV caches into every branch, they pass findings between agents as text…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-11-06/octopus-paper-shares-one-kv-cache-across-sub-agents-and-lift (subscription information: https://polylog.news/pricing).