{"slug": "tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang", "title": "Tsinghua AI Open-Sources AKV, a Cache Engineering Server for LLM Agents on SGLang", "summary": "Tsinghua researchers open-sourced AKV, an SGLang-based serving system that lets LLM agents drop and reposition KV cache instead of summarizing conversation history, cutting average cost by 27.7% across 4 agent benchmarks. The release targets teams running agent workloads on SGLang who want lower serving costs without history summarization.", "body_md": "Tsinghua researchers released AKV, an SGLang-based serving system for cache engineering. LLM agents drop and reposition KV cache instead of summarizing history, cutting average cost 27.7% across 4 agent benchmarks.\n\nThe post [Tsinghua AI Open-Sources AKV, a Cache Engineering Server for LLM Agents on SGLang](https://www.marktechpost.com/2026/10/26/tsinghua-akv-kv-cache-engineering-llm-agents/) appeared first on [MarkTechPost](https://www.marktechpost.com).", "url": "https://wpnews.pro/news/tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang", "canonical_source": "https://www.marktechpost.com/2026/10/26/tsinghua-akv-kv-cache-engineering-llm-agents/", "published_at": "2026-10-26 07:56:45+00:00", "updated_at": "2026-10-26 08:15:16.656540+00:00", "lang": "en", "topics": ["ai-agents", "large-language-models", "ai-infrastructure", "ai-tools", "mlops"], "entities": ["Tsinghua University", "AKV", "SGLang"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang", "markdown": "https://wpnews.pro/news/tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang.md", "text": "https://wpnews.pro/news/tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang.txt", "jsonld": "https://wpnews.pro/news/tsinghua-ai-open-sources-akv-a-cache-engineering-server-for-llm-agents-on-sglang.jsonld"}}