Ollaya – Ollama for open-source, Jev-style decision models
Ollaya – Ollama for open-source, Jev-style decision models
Ollaya – 面向开源、Jev 风格决策模型的 Ollama
Run decision models locally. Ask typed questions about any text or JSON and get calibrated answers in milliseconds. Private, open source, on your own hardware. 在本地运行决策模型。针对任何文本或 JSON 提出结构化问题，并在毫秒内获得经过校准的答案。私有、开源，运行在您自己的硬件上。
Fast
极速
Decisions in milliseconds. A decision model answers in a single forward pass, with no token-by-token generation. On your own GPU, a five-question request to Laya takes about 10 ms, end to end through the HTTP API. 决策仅需毫秒级时间。决策模型通过单次前向传播即可回答，无需逐个 token 生成。在您自己的 GPU 上，通过 HTTP API 对 Laya 发起包含五个问题的请求，端到端耗时仅约 10 毫秒。
Drop-in compatible
无缝兼容
Speaks TypeSafe’s API. Ollaya serves /v1/systemone and /v1/models with TypeSafe’s request and response shapes. The official TypeSafe Python SDK 0.7.1 works unchanged against a local server.
支持 TypeSafe API。Ollaya 提供与 TypeSafe 请求和响应格式一致的 /v1/systemone 和 /v1/models 接口。官方 TypeSafe Python SDK 0.7.1 无需修改即可直接连接本地服务器。
Open models
开源模型
Open weights, ready to pull. Start with Laya from Convai Innovations: an English model, a 100+ language model, a model fine-tuned for typed decisions, and a router that picks for you. 开放权重，随时可拉取。从 Convai Innovations 的 Laya 模型开始：包括英语模型、支持 100 多种语言的模型、针对结构化决策微调的模型，以及为您自动选择模型的路由。
- laya: Open decision models from Convai Innovations. Typed, calibrated answers to choice, score and yes/no questions in a single forward pass, in English and 100+ languages. laya: 来自 Convai Innovations 的开源决策模型。通过单次前向传播，以英语和 100 多种语言提供针对选择、评分和是非题的结构化、已校准答案。
- decider: Decoder decision models by Mapika on Qwen3.5: the answer is read from option-letter logits in one forward pass. The most accurate open decision model Ollaya ships. decider: 由 Mapika 基于 Qwen3.5 开发的解码器决策模型：通过单次前向传播从选项字母的 Logits 中读取答案。这是 Ollaya 目前发布的最精确的开源决策模型。
- nli: Zero-shot classifiers by Moritz Laurer: every option becomes a hypothesis scored for entailment. The most accurate encoder model on typed decisions in our tests. nli: 由 Moritz Laurer 开发的零样本分类器：每个选项都成为一个用于蕴含评分的假设。在我们针对结构化决策的测试中，这是最精确的编码器模型。
- gliclass: Instruction-following zero-shot classifier by Knowledgator: all options of a question are scored in one pass, so cost barely grows with the number of options. gliclass: 由 Knowledgator 开发的指令遵循零样本分类器：单次传递即可对问题的所有选项进行评分，因此计算成本几乎不会随选项数量增加。
- qwen3guard: Safety guard by the Qwen team: is a text safe, controversial or unsafe, and which unsafe category? It answers its own built-in questions, in 119 languages, in one forward pass. qwen3guard: 由 Qwen 团队开发的安全防护模型：判断文本是否安全、有争议或不安全，并指出具体的不安全类别。它能在单次前向传播中以 119 种语言回答其内置的安全问题。
- kev: Decision models by Jared Palmer: a LoRA on a Qwen3.5 base plus a pointer head that scores every option at its own span, in one forward pass per question. Calibrated with Kev’s own temperature. kev: 由 Jared Palmer 开发的决策模型：基于 Qwen3.5 的 LoRA 模型，外加一个指针头，在每个问题的一次前向传播中对每个选项的跨度进行评分。使用 Kev 自定义的温度参数进行校准。
- von: Decision model by Victor Hugo Panisa on ModernBERT-large: every option is scored at its own marker, all options of a question in one pass, with an input-conditioned calibration. 8k-token context. von: 由 Victor Hugo Panisa 基于 ModernBERT-large 开发的决策模型：每个选项在其标记处进行评分，单次传递即可处理所有选项，并具备输入条件校准功能。支持 8k token 上下文。
Your data stays yours
数据归您所有
- Private by default: Tickets, emails and user messages are often the most sensitive data you have. With Ollaya they are scored where they already live. 默认私有: 工单、电子邮件和用户消息通常是您最敏感的数据。使用 Ollaya，这些数据在它们原本存储的地方即可完成评分。
- Local: Runs on your machine with ONNX Runtime, on the CPU or an NVIDIA GPU. The server listens on 127.0.0.1 by default. 本地运行: 使用 ONNX Runtime 在您的机器上运行，支持 CPU 或 NVIDIA GPU。服务器默认监听 127.0.0.1。
- Open weights: Weights come from their authors’ Hugging Face repositories, pinned to a commit and checked against sha256. Ollaya never re-hosts them, and the runtime is Apache-2.0. 开放权重: 权重来自作者的 Hugging Face 仓库，锁定特定提交版本并经过 sha256 校验。Ollaya 从不重新托管这些权重，且运行时采用 Apache-2.0 协议。
- No per-token fees: Run as many decisions as your hardware can handle. No metering and no API bill. 无按 Token 收费: 在硬件承载范围内运行任意数量的决策任务。无需计量，没有 API 账单。
Platforms
平台支持
Runs where you work. A desktop app and a command line for macOS, Windows and Linux, and a Docker image for servers. Every model runs on the CPU; an NVIDIA GPU on Linux, in WSL 2 or in Docker takes a request down to milliseconds. 随处运行。提供适用于 macOS、Windows 和 Linux 的桌面应用和命令行工具，以及用于服务器的 Docker 镜像。所有模型均可在 CPU 上运行；在 Linux、WSL 2 或 Docker 中使用 NVIDIA GPU 可将请求响应时间缩短至毫秒级。
Get up and running in minutes.
几分钟内即可上手。
One binary, one command: ollaya run laya.
一个二进制文件，一条命令：ollaya run laya。