{"id":"1b0db055-6314-4099-947b-f369f7b42783","kind":"companies","code":"UB-C12947","name":"Shiliu","nameBasis":"translated","summary":"Let Agents remember how users change","description":"Shiliu (official domain shiliu.me) is long-term context infrastructure for AI apps: cross-session memory, change tracking, evidence tracing, on-demand recall, and memory governance. Published LongMemEval-CN Chinese 500-question per-item results; first round 499/500.","country":null,"category":null,"originalName":"时流","eventDate":null,"website":"https://shiliu.me/","sourceUpdatedAt":"2026-10-26T00:00:00.000Z","englishUpdatedAt":"2026-11-01T02:12:46.572Z","reviewedAt":"2026-11-01T02:12:46.572Z","snapshotAt":"2026-11-01T00:40:31.077Z","sourceHash":"09986e0c1d9a34af22c9e0bd6a6a4c9667fabcf758dc84f657673ddf803953f2","companyId":null,"foundedYear":null,"logoUrl":null,"originalText":"让 Agent 记得用户如何变化\n\n时流 Shiliu（官方域名 shiliu.me）是面向 AI 应用的长期上下文基础设施，提供跨会话记忆、变化追踪、证据追溯、按需召回与记忆治理。已公开 LongMemEval-CN 中文500题逐题结果，首轮499/500。","sources":[],"related":[{"id":"98abf93d-f22f-4226-b0d0-3c5f79bfe2e0","kind":"products","code":"UB-D6945","name":"Shiliu","nameBasis":"transliterated","summary":"Shiliu focuses on cross-model, cross-agent Chinese long-term memory infrastructure, covering seven-path memory recall, temporal reasoning, conflict detection, expired-memory cleanup, context compression, and change provenance. The team led a unified memory API and SDK that closes the loop on write, recall, correction, and cross-agent sharing. On the LongMemEval-CN Chinese long-memory benchmark it passed 499/500 items in the first round (99.8% accuracy) and achieved 82.3% input-token compression in staged experiments. The product is now in deliverable Alpha, with public API, permission isolation, audit trails, and enterprise private deployment.","description":"Shiliu focuses on cross-model, cross-agent Chinese long-term memory infrastructure, covering seven-path memory recall, temporal reasoning, conflict detection, expired-memory cleanup, context compression, and change provenance. The team led a unified memory API and SDK that closes the loop on write, recall, correction, and cross-agent sharing. On the LongMemEval-CN Chinese long-memory benchmark it passed 499/500 items in the first round (99.8% accuracy) and achieved 82.3% input-token compression in staged experiments. The product is now in deliverable Alpha, with public API, permission isolation, audit trails, and enterprise private deployment.","country":null,"category":"General-purpose agents","originalName":"时流","eventDate":null,"website":null,"sourceUpdatedAt":"2026-10-26T21:00:15.342Z","englishUpdatedAt":"2026-10-28T00:56:07.347Z","reviewedAt":"2026-10-28T00:56:07.347Z","snapshotAt":"2026-10-28T00:47:51.550Z","sourceHash":"b3c58cdeb89c679a9812310933a4ea1e79bd17d1530c77c80ada54d6b5a3549b","companyId":"1b0db055-6314-4099-947b-f369f7b42783","foundedYear":null,"logoUrl":null,"path":"/products/UB-D6945"}],"finance":{"funding":{"metric":"funding","count":0,"latest":null,"years":[],"points":[],"linePoints":[],"conflictGroups":0,"omittedRecordCount":0,"records":[],"totalRaised":null},"valuation":{"metric":"valuation","count":0,"latest":null,"years":[],"points":[],"linePoints":[],"conflictGroups":0,"omittedRecordCount":0,"records":[]},"revenue":{"metric":"revenue","count":0,"latest":null,"years":[],"points":[],"linePoints":[],"conflictGroups":0,"omittedRecordCount":0,"records":[]}},"events":[],"team":[],"productsRelated":[{"id":"98abf93d-f22f-4226-b0d0-3c5f79bfe2e0","kind":"products","code":"UB-D6945","name":"Shiliu","nameBasis":"transliterated","summary":"Shiliu focuses on cross-model, cross-agent Chinese long-term memory infrastructure, covering seven-path memory recall, temporal reasoning, conflict detection, expired-memory cleanup, context compression, and change provenance. The team led a unified memory API and SDK that closes the loop on write, recall, correction, and cross-agent sharing. On the LongMemEval-CN Chinese long-memory benchmark it passed 499/500 items in the first round (99.8% accuracy) and achieved 82.3% input-token compression in staged experiments. The product is now in deliverable Alpha, with public API, permission isolation, audit trails, and enterprise private deployment.","description":"Shiliu focuses on cross-model, cross-agent Chinese long-term memory infrastructure, covering seven-path memory recall, temporal reasoning, conflict detection, expired-memory cleanup, context compression, and change provenance. The team led a unified memory API and SDK that closes the loop on write, recall, correction, and cross-agent sharing. On the LongMemEval-CN Chinese long-memory benchmark it passed 499/500 items in the first round (99.8% accuracy) and achieved 82.3% input-token compression in staged experiments. The product is now in deliverable Alpha, with public API, permission isolation, audit trails, and enterprise private deployment.","country":null,"category":"General-purpose agents","originalName":"时流","eventDate":null,"website":null,"sourceUpdatedAt":"2026-10-26T21:00:15.342Z","englishUpdatedAt":"2026-10-28T00:56:07.347Z","reviewedAt":"2026-10-28T00:56:07.347Z","snapshotAt":"2026-10-28T00:47:51.550Z","sourceHash":"b3c58cdeb89c679a9812310933a4ea1e79bd17d1530c77c80ada54d6b5a3549b","companyId":"1b0db055-6314-4099-947b-f369f7b42783","foundedYear":null,"logoUrl":null,"path":"/products/UB-D6945"}],"investors":[],"path":"/companies/UB-C12947","structured":{"city":null,"country":null,"location":null,"role":null,"category":null,"company":null},"definition":"Shiliu is a company.","summaryBasis":"translated","thin":false,"canonical":"https://uniquebase.ai/companies/UB-C12947"}