Simon Willison 发布 OpenAI Decisions API
llm-openai-decisions 0.1a0
Release: llm-openai-decisions 0.1a0
发布：llm-openai-decisions 0.1a0
OpenAI released their new Jev-style Decisions API, as previously announced at last week's DevDay.
OpenAI 发布了其全新的 Jev 风格 Decisions API，正如上周 DevDay 上此前宣布的那样。
Since I already have an llm-typesafe plugin for talking to Jev, I had GPT-6 Astra read the new OpenAI API documentation and build an llm-openai-decisions plugin inspired by llm-typesafe.
由于我已经有一个用于与 Jev 通信的 llm-typesafe 插件，我便让 GPT-6 Astra 阅读新的 OpenAI API 文档，并构建了一个受 llm-typesafe 启发的 llm-openai-decisions 插件。
Unlike Jev, the new gpt-6-luna decision model supports image input in addition to text. Both models charge for input it and not for output: OpenAI's is 10 cents per million input tokens, Jev's is 4.2 cents per million.
与 Jev 不同，新的 gpt-6-luna 决策模型除了支持文本输入外，还支持图像输入。两个模型都对输入收费而不对外部输出收费：OpenAI 的费用是每百万输入 token 10 美分，Jev 的费用是每百万输入 token 4.2 美分。
Otherwise the API shape is very similar to Jev, at least conceptually. Jev supports three question types for yes/no, choices, or scores. OpenAI Decisions supports the same three types.
否则，API 的结构非常相似，至少在概念上如此。Jev 支持三种问题类型：是/否、选择或评分。OpenAI Decisions 也支持相同的三种类型。
Install the plugin like this:
像这样安装插件：
llm install llm-openai-decisionsHere's an example query against an image attachment:
以下是一个针对图像附件的查询示例：
llm -m openai-decisions/gpt-6-luna \
-a https://static.simonwillison.net/static/2025/two-pelicans.jpg \
-s 'Does this image contain any mammals?'And example output:
以及示例输出：
{"type": "predicate", "name": "evaluation", "probability": 0.0}Consult the README for full details of how to run the other types of questions.
请参阅 README 以获取运行其他类型问题的完整详情。
Tags: openai, llm, coding-agents, jev
标签：openai, llm, coding-agents, jev
更进一步：量化金融体系
看懂新闻只是起点——沿量化金融路径，把它变成能交付的工程能力