corpus.blog
Claim your blog
Record · corpus.blog/posts/01a08789-72f4-73c1-b9b5-f027cfb384ce

Subagents Account for Most Token Costs in Long Agent Runs: Fixes That Cut Usage 70 to 90 Percent in Practice

jackmaguire.org · published 2 June 2026

A record is what was published and who pointed at it. The text of the post is not held here: read it at the source, or at its Wayback capture.

Read the post at jackmaguire.org Wayback capture
0blogs citing
First cited
—
Most recent
—
Rank this month
not ranked
External links
32
Words captured
3,145

Cited by

No blog we hold has cited this post yet.

Links from this post

Link
Host
https://www.reddit.com/r/AI_Agents/
reddit.com
https://www.reddit.com/r/ClaudeCode/?ref=endnotes.net
reddit.com
https://news.ycombinator.com/
news.ycombinator.com
https://arxiv.org/abs/2604.22750
arxiv.org
https://arxiv.org/pdf/2604.22750
arxiv.org
https://longjubai.github.io/agent_token_consumption
longjubai.github.io
https://digitaleconomy.stanford.edu/news/how-are-ai-agents-spending-your-tokens/
digitaleconomy.stanford.edu
https://www.zdnet.com/article/your-cost-for-ai-agents-will-be-wildly-variable-and-unpredictable/
zdnet.com
https://github.com/microsoft/LLMLingua?ref=ashu.co
github.com
https://www.microsoft.com/en-us/research/blog/llmlingua-innovating-llm-efficiency-with-prompt-compression/
microsoft.com
https://arxiv.org/abs/2403.12968
arxiv.org
https://www.speakeasy.com/blog/how-we-reduced-token-usage-by-100x-dynamic-toolsets-v2
speakeasy.com
https://www.speakeasy.com/blog/100x-token-reduction-dynamic-toolsets
speakeasy.com
https://www.zyte.com/blog/why-ai-agents-struggle-with-web-scraping/
zyte.com
https://www.zyte.com/blog/why-10-million-tokens-wont-save-your-ai-agent/
zyte.com
https://docs.anthropic.com/en/docs/build-with-claude/prompt-caching
docs.anthropic.com
Code execution with MCP: Building more efficient agents Read at the source ↗
anthropic.com
https://www.anthropic.com/engineering/effective-context-engineering-for-ai-agents
anthropic.com
https://arxiv.org/pdf/2603.27277
arxiv.org
https://dev.to/deusdata/how-i-cut-my-ai-coding-agents-token-usage-by-120x-with-a-code-knowledge-graph-4a3d
dev.to
https://github.com/etr/wonk
github.com
https://news.ycombinator.com/item?id=48159227
news.ycombinator.com
https://arxiv.org/abs/2210.03629
arxiv.org
https://tianpan.co/blog/2026-04-16-retry-budget-llm-agent-cost-amplification
tianpan.co
https://sureprompts.com/blog/context-engineering-the-2026-replacement-for-prompt-engineering
sureprompts.com
Build AI agents with short-term & long-term memory in Redis Read at the source ↗
redis.io
https://github.com/redis/agent-memory-server
github.com
https://techcommunity.microsoft.com/blog/azure-managed-redis/supercharging-ai-agents-with-memory-on-azure-managed-redis/4457407
techcommunity.microsoft.com
https://www.anthropic.com/engineering/advanced-tool-use
anthropic.com
https://github.com/ryoppippi/ccusage/issues/950
github.com
https://www.zyte.com/blog/agentic-web-scraping/
zyte.com
https://arxiv.org/abs/2605.20049
arxiv.org
Elsewhere on jackmaguire.org (3)