PulseAugur
实时 12:06:28
English(EN) I burned 603M tokens in seven days without knowing where they went. Hermes Agent's auxiliary block was silently firing background tasks through kimi-k2.6 (1T pa

AI 代理后台任务消耗 6.03 亿 token;开发者实施路由

一位 AI 开发者发现,由于后台任务静默运行,他们的 Hermes Agent 在七天内消耗了大量 token,总计 6.03 亿个。问题追溯到 kimi-k2.6 模型。开发者实施了显式路由来优化 token 使用,将不同任务分配给更轻量级或更合适的模型,如 rnj-1:8bgemma3:12bdeepseek-v4-flashkimi-k2.5,从而将成本降低了高达 125 倍。 AI

影响 优化 LLM 路由可以显著降低 AI 应用的运营成本并提高效率。

排序理由 该集群描述了用户级别的 AI 代理资源消耗优化和修复,而非新的模型发布或重大行业事件。

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。 我们如何撰写摘要 →

AI 代理后台任务消耗 6.03 亿 token;开发者实施路由

本文如何被排名

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Tool
该集群描述了用户级别的 AI 代理资源消耗优化和修复,而非新的模型发布或重大行业事件。
Source corroboration
Single-source cluster
Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics
product, infra
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
100 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

完整方法见我们的编辑标准

报道来源 [1]

  1. Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] ·

    七天内我消耗了6.03亿个token却不知道去向。Hermes Agent的辅助模块通过kimi-k2.6(1T pa)在后台静默运行任务。

    I burned 603M tokens in seven days without knowing where they went. Hermes Agent's auxiliary block was silently firing background tasks through kimi-k2.6 (1T params). The fix: explicit routing. • Titles/search/skills → rnj-1:8b (125x lighter) • Classification → gemma3:12b (12B) •…