Sieve Transparent context reduction for LLMs. https://llmsieve.dev/ https://github.com/llmsieve/llm-sieve en Mon, 15 Jun 2026 13:23:14 -0000 Mon, 15 Jun 2026 13:23:14 -0000 1440 MkDocs RSS plugin - v1.19.0 None Sieve https://llmsieve.dev/ The two causes of your token bill Sieve Engineering Design Agent token bills have two separate causes. Why compression and reduction are complementary fixes, not rivals. https://llmsieve.dev/blog/2026/06/15/the-two-causes-of-your-token-bill/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/15/the-two-causes-of-your-token-bill/ Sieve, mem0, Zep: three shapes of agent memory Sieve Engineering Design mem0's SDK, Zep's managed platform, and Sieve's transparent proxy solve agent memory in three different shapes. How to pick. https://llmsieve.dev/blog/2026/06/10/sieve-mem0-zep-three-shapes-of-agent-memory/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/10/sieve-mem0-zep-three-shapes-of-agent-memory/ Persistent memory for Ollama, in about five minutes Sieve Engineering Guides Give any Ollama model persistent, encrypted memory in about five minutes — without changing your client code. https://llmsieve.dev/blog/2026/06/10/persistent-memory-for-ollama-in-about-five-minutes/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/10/persistent-memory-for-ollama-in-about-five-minutes/ The hidden cost of context Sieve Engineering Engineering Why LLM agent token bills grow faster than developers expect — and what context actually costs in compute at scale. https://llmsieve.dev/blog/2026/06/10/the-hidden-cost-of-context/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/10/the-hidden-cost-of-context/ What always-on agents stand to gain from a context proxy Sieve Engineering Design OpenClaw, Hermes, and the always-on agent workload — why gateway-style assistants are the heaviest context spenders yet. https://llmsieve.dev/blog/2026/06/10/what-always-on-agents-stand-to-gain-from-a-context-proxy/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/10/what-always-on-agents-stand-to-gain-from-a-context-proxy/ Why Sieve Sieve Engineering Design Why Sieve is a transparent proxy rather than a memory library or prompt compaction — and what that trade-off costs you. https://llmsieve.dev/blog/2026/06/09/why-sieve/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/09/why-sieve/ Compute is the bottleneck. Tokens are just the price tag. Sieve Engineering Economics Compute is the binding constraint on AI. Tokens just reprice it. Workload shape is the one lever you fully control. https://llmsieve.dev/blog/2026/06/09/compute-is-the-bottleneck-tokens-are-just-the-price-tag/ Mon, 15 Jun 2026 13:23:15 +0000 Sieve https://llmsieve.dev/blog/2026/06/09/compute-is-the-bottleneck-tokens-are-just-the-price-tag/