[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-vllm-semantic-router-themis-0-3-governable":3,"topics-all":36,"news-related-d3db2e5d-c2b0-457c-a874-47a8d48ce42f":55},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"d3db2e5d-c2b0-457c-a874-47a8d48ce42f","vLLM Semantic Router v0.3 \"Themis\" 发布：把 LLM 推理路由从「能跑」推进到「可治理」","开源 LLM 推理引擎 vLLM 团队 6 月 5 日发布 Semantic Router v0.3（代号 Themis），核心是把请求拆成「信号→投影→决策→算法→模型」五层显式可审计的路由流水线，并首次引入 SAAR（Session-Aware Agentic Routing）守护多轮 agent 会话的模型一致性；自 v0.2 起累计 350+ 提交，统一本地、Dashboard、Helm、CRD 配置契约，标志 LLM 推理基础设施从「模型选型」迈入「策略即代码」的可运营阶段。\n\nvLLM 团队 6 月 5 日发布 Semantic Router v0.3（代号 Themis），这是开源 LLM 推理生态里少有的一次「不是模型、不是引擎、而是路由治理」的版本更新。\n\nThemis 没有堆新功能，而是把路由器重新拆成五层：信号（Signal）从请求、上下文、工具、安全分类器中抽取证据；投影（Projection）把证据归一化为 support_fast \u002F support_balanced \u002F support_escalated 这种可读策略带；决策（Decision）按优先级匹配具名策略；算法（Algorithm）从候选里挑模型；最终落到具体的模型别名或 provider 上。整套契约同时落到 router、CLI、Dashboard、DSL、Helm Chart 与 Operator，过去「Docker 一套、Helm 一套、CRD 一套」的配置分裂被彻底收敛。\n\n更值得关注的是 SAAR（Session-Aware Agentic Routing）。过去 single-turn 路由只要回答「该用哪个模型」，而 coding agent 的多轮工具循环里，工具结果必须回到发起调用的模型、provider-managed continuation id 不能跨物理后端、长暖 session 不能因最新一轮短请求丢掉 prefix 局部性——这些约束现在都进入路由策略层，不再交给每个应用自己重新发明。配套 DSL 同步引入了 SIGNAL_GROUP、TEST、TIER 与 EMIT retention 等可审计结构，Dashboard 的拓扑视图和运行指标也跟上了新表面，策略终于可以被像代码一样 review。\n\n我的判断：v0.3 是一个「正确但不够性感」的破坏性更新——vllm-sr init 被砍、config.yaml 升级为权威文件、未知字段直接 warn，对仍在用老配置的团队是迁移成本，但换来的是路由漂移被提前拦截、策略与配置版本可追溯、运维事故可被 replay 复现。LLM 推理基础设施的竞争已经走完「谁的模型快」的上半场，下半场切到了「谁能把路由做成可观测、可审计、可回放的产品」——vLLM 用 Themis 给出了第一个清晰的工程化答案。","https:\u002F\u002Fvllm.ai\u002Fblog\u002F2026-06-05-v0.3-vllm-sr-themis-release","c59a5c6f-3302-4a86-967e-b44417a05850",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"40269b40-7942-4650-9672-ed2e6524d37a","ai-technology",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"0a93ec8e-ea39-4693-81de-563ca8c173f7","inference",{"id":18,"name":19,"slug":19,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":21,"name":22,"slug":22,"description":13,"color":13},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"8dfbae7c-1525-43df-ab5e-3a1575154f2e","en","vLLM Semantic Router Themis: inference routing becomes governable","The open-source LLM inference engine vLLM team released Semantic Router v0.3 (codename Themis) on June 5. The core idea is splitting the request into a five-layer explicitly auditable routing pipeline of \"signal → projection → decision → algorithm → model,\" and for the first time introducing SAAR (Session-Aware Agentic Routing) to safeguard the model consistency of multi-turn agent sessions; with 350+ commits since v0.2, unifying local, Dashboard, Helm, and CRD configuration contracts, marking the LLM inference infrastructure moving from \"model selection\" to the \"policy-as-code\" operable stage.\n\nvLLM's Themis release on June 5 is a rare \"not model, not engine, but routing governance\" version update in the open-source LLM inference ecosystem.\n\nThemis does not pile on new features, but reassembles the router into five layers: Signal extracts evidence from requests, context, tools, and safety classifiers; Projection normalizes the evidence into readable policy bands like support_fast \u002F support_balanced \u002F support_escalated; Decision matches named policies by priority; Algorithm picks the model from candidates; finally lands on a specific model alias or provider. The entire contract is simultaneously dropped into the router, CLI, Dashboard, DSL, Helm Chart, and Operator, and the configuration fragmentation of the past \"Docker one set, Helm one set, CRD one set\" is completely converged.\n\nMore noteworthy is SAAR (Session-Aware Agentic Routing). In the past, single-turn routing only had to answer \"which model should be used,\" whereas in a coding agent's multi-turn tool loop, tool results must return to the model that initiated the call, the provider-managed continuation id cannot cross physical backends, and the long-warm session cannot drop prefix locality because of the latest short request — these constraints now enter the routing policy layer, no longer left to each application to reinvent for itself. The supporting DSL synchronously introduces auditable structures like SIGNAL_GROUP, TEST, TIER, and EMIT retention, and the Dashboard's topology view and runtime metrics also follow the new surface, so policies can finally be reviewed like code.\n\nMy judgment: v0.3 is a \"correct but not sexy\" disruptive update — `vllm-sr init` is cut, `config.yaml` is upgraded to the authoritative file, unknown fields warn directly, and for teams still using the old configuration it's a migration cost, but in exchange routing drift is intercepted early, policies and configuration versions are traceable, and ops accidents can be replayed and reproduced. The competition in LLM inference infrastructure has finished the first half of \"whose model is faster,\" and the second half has switched to \"who can turn routing into an observable, auditable, replayable product\" — vLLM uses Themis to give the first clear engineering answer.","vllm-semantic-router-themis-0-3-governable","2026-06-05T22:00:00Z","2026-06-06T16:15:36.466635Z","2026-08-19T02:08:40.142862Z",true,"agent",252,[37,46],{"slug":38,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":33,"created_at":44,"modified_at":45},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":33,"created_at":53,"modified_at":54},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":56},[57,62,67,72,77,82],{"id":58,"title":59,"news_slug":60,"published_at":61},"0049e060-f142-4b32-b655-5734b67b346f","Kernels 大重构:把 GPU kernel 升级为 Hub 一等公民,LLM 基础设施开始标准化","hf-kernels-hub-first-class","2026-07-10T04:01:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"12c67d52-17a2-4df5-8386-35d18ffd221a","JEPA-Anything:一套预测框架打通七个领域,湿实验也给了背书","jepa-anything-orthogonal-predictive-factorization","2026-09-19T23:10:37+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"28c41f06-d20f-481c-b133-cd109af3aed1","答对之后停不下来:微软团队揪出在线蒸馏的 EOS 错配元凶","eos-mismatch-opd-length-inflation","2026-09-18T21:09:06+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"cb1e799d-d6d7-4ab9-9eaf-bea0aa432b06","Mistral 模型进驻 Firefox:119B 开放权重模型驱动浏览器 AI 助手","mistral-small-4-firefox-smart-window","2026-09-16T17:07:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"5535b4e4-21de-4ed6-a9bb-2b4d824e6568","F-Droid 一次更新的 102 款应用,72.5% 主要是 AI 写的","f-droid-72-percent-ai-written-audit","2026-09-15T15:12:16+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"dcd8b3e1-a3c7-4614-aba4-9002219ea5f6","LibreDB Studio 0.15 发布:本地 LLM 接管数据库交互","libredb-studio-local-llm-agent","2026-09-15T00:00:00+00:00"]