Tag
Trigger.dev's new chat agent lets developers build durable, stateful AI chat experiences that survive refreshes and crashes, with no timeouts and the ability to pause for permission before risky tool calls.
TokTier is a stateful tokenization service that cuts tokenization overhead in agentic LLM serving by reusing stable token boundaries and running exact CPU/GPU tokenization, reducing time-to-first-token by 16–34% under vLLM.
This article explores the tradeoffs between stateful and stateless agent design for scalable AI systems, providing implementation examples using Groq API and Llama 3.1 8B Instant model.
Introduces ContinuityBench, a benchmark and systems study for stateful failover in multi-provider LLM routing, proposing new metrics (CPR, CLO) and a history-forwarding proxy architecture achieving 99.20% context preservation.
RecGPT-V3 introduces a stateful, hybrid-modal recommender that uses a Memory Hub and Latent Intent Reasoning to cut computation by 55.8% and output token cost by 200x, achieving significant gains in Taobao's feed.
This article provides a detailed breakdown of the Hermes Agent architecture, including the event bus, adapter layer, GatewayRunner core scheduler layer, etc., revealing the design pattern of the Agent as a stateful, event-driven runtime.
Proposes SKG-Eval, a quasi-deterministic evaluation framework for multi-turn dialogue that uses incremental semantic knowledge graphs to detect cross-turn inconsistencies, contradiction, and topic drift, achieving higher correlation with human judgments.