bailian-docs-llm-wiki
阿里云百炼平台技术文档知识库(LLM Wiki)。当用户查询百炼模型列表、API 参数、错误码、 应用开发(智能体/RAG/知识库/记忆/插件)、模型对比与定价、SDK/OpenAI 兼容接口、 多模态能力(语音/图片/视频)、Token 计费等百炼相关问题时激活。 包含 models 结构化模型市场数据(含…
Skill by modelstudioai
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/modelstudioai-bailian-docs-llm-wiki
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: backend, ml
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md and code. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 18
alibabacloud.com· found in codealiyun.com· found in codecdn.translate.alibaba.com· found in codecdn.wanx.aliyuncs.com· found in codecodeload.github.com· found in install commanddashscope.aliyuncs.com· found in codedashscope.oss-cn-beijing.aliyuncs.com· found in codedocs.bailian.console.aliyun.com· found in codeg-adoc.alcasset.com· found in codehelp-static-aliyun-doc.aliyuncs.com· found in codehelp.aliyun.com· found in codeimg.alicdn.com· found in codellm-19hapw1e7htdf9lh.cn-beijing.maas.aliyuncs.com· found in codemaas-infra-public.oss-cn-hangzhou.aliyuncs.com· found in codesynthesis-source.oss-accelerate.aliyuncs.com· found in codevigen-video.oss-cn-shanghai.aliyuncs.com· found in codewanx.alicdn.com· found in codexxx.wav· found in code
Files it touches
~/.claude/skills/bailian-docs-llm-wiki
What's inside
- Skill name
bailian-docs-llm-wiki- Description
- 阿里云百炼平台技术文档知识库(LLM Wiki)。当用户查询百炼模型列表、API 参数、错误码、 应用开发(智能体/RAG/知识库/记忆/插件)、模型对比与定价、SDK/OpenAI 兼容接口、 多模态能力(语音/图片/视频)、Token 计费等百炼相关问题时激活。 包含 models 结构化模型市场数据(含 contextWindow/QPM/价格/sample code)、 wiki 合成层(主题页/概念页/对比页)和 raw 原始文档层; 模型规格类问题先查 models/index.md,文档类问题先查 wiki/index.md。
- License
- Apache-2.0
- Instructions
- 172 lines in SKILL.md · view on GitHub
Files (200)
- README.md
- SKILL.md
- llms.txt
- families.jsonl · models/
- Kimi-K2.json · models/groups/
- MiniMax-M2.1.json · models/groups/
- MiniMax-speech-market-place.json · models/groups/
- MiniMax-video-market-place.json · models/groups/
- cosyvoice.json · models/groups/
- decision-model.json · models/groups/
- deepseek.json · models/groups/
- embedding.json · models/groups/
- fun-asr-flash.json · models/groups/
- fun-asr-realtime.json · models/groups/
- fun-asr.json · models/groups/
- glm-4.5.json · models/groups/
- glm-fast.json · models/groups/
- gui-plus.json · models/groups/
- happyhorse-1.0.json · models/groups/
- happyhorse-1.1.json · models/groups/
- happyoyster-1.0.json · models/groups/
- image-erase-completion.json · models/groups/
- image-instance-segmentation.json · models/groups/
- kimi-models-market-place.json · models/groups/
- kling-models-market-place.json · models/groups/
- minimax-models-market-place.json · models/groups/
- paraformer-8k-v2.json · models/groups/
- paraformer-realtime-8k-v2.json · models/groups/
- paraformer-realtime-v2.json · models/groups/
- paraformer-v2.json · models/groups/
- pixverse-c1-market-place.json · models/groups/
- pixverse-capability-market-place.json · models/groups/
- pixverse-market-place.json · models/groups/
- pixverse-v6-market-place.json · models/groups/
- qwen-audio-3.0-asr.json · models/groups/
- qwen-audio-3.0-realtime.json · models/groups/
- qwen-audio-3.0-tts.json · models/groups/
- qwen-audio-3.1-asr-message.json · models/groups/
- qwen-audio-3.1-asr.json · models/groups/
- qwen-audio-3.1-realtime.json · models/groups/
- qwen-audio-3.1-tts-next.json · models/groups/
- qwen-audio-3.1-tts.json · models/groups/
- qwen-audio-asr-flash-filetrans.json · models/groups/
- qwen-audio-asr-flash-streaming.json · models/groups/
- qwen-deep-research.json · models/groups/
- qwen-doc-turbo.json · models/groups/
- qwen-embedding.json · models/groups/
- qwen-flash-character.json · models/groups/
- qwen-flash.json · models/groups/
- qwen-image-2.0.json · models/groups/
- qwen-image-2.1.json · models/groups/
- qwen-image-3.0.json · models/groups/
- qwen-long.json · models/groups/
- qwen-max.json · models/groups/
- qwen-mt-flash.json · models/groups/
- qwen-mt-image.json · models/groups/
- qwen-mt-lite.json · models/groups/
- qwen-mt-plus.json · models/groups/
- qwen-mt-uni.json · models/groups/
- qwen-plus-character.json · models/groups/
- …and 140 more
SKILL.md (first lines)
# 百炼文档知识库
> 阿里云百炼平台的完整技术文档知识库,涵盖模型使用、应用开发、API 参考等内容。
## 何时使用
当用户涉及以下场景时激活此 Skill:
- 查询百炼平台的模型列表、模型参数、调用方式(含具体模型的 contextWindow、QPM、定价、sample code 等结构化字段 → 查 `models/`)
- 查阅百炼 API 参数、请求/响应格式、错误码
- 了解百炼应用开发(智能体、RAG、知识库、记忆、插件等)
- 选择模型、比较模型能力、了解模型定价与限流
- 使用百炼 SDK / OpenAI 兼容接口
- 语音识别、语音合成、图片生成、视频生成等多模态能力
- Token Plan、计费、免费额度等商务问题
## 何时**不要**使用
下面这些场景与百炼文档无关,**不要激活**本 Skill:
- 询问 OpenAI / Anthropic / Google 等其他厂商的 API、模型或定价
- 通用编程问题、框架问题(React、Vue、Spring 等)
- `bl` CLI 命令本身的用法(那是 `bailian-cli` skill 的职责)
- 与百炼无关的阿里云产品(OSS、ECS、RDS 等)
## 文档层级
### 1. Wiki 层(合成层) — `wiki/`
由 LLM 自动合成的结构化 Wiki,包含三类页面:
- **主题页**:按功能领域聚合的综合文档(`wiki/guides/*.md`、`wiki/api/*.md`)
- **概念页**:跨主题的横切概念(`wiki/concepts/*.md`,如 RAG、流式输出、Function Calling 等)
- **对比页**:同类方案的结构化对比分析(`wiki/comparisons/*.md`,含对比表格)
**优先查阅 `wiki/index.md`,它是全部 wiki 页面的索引入口。**
每篇主题页末尾都有 `## 来源文档` 段落,列出对应的 raw 原文路径,可用于追溯和深度阅读。
### 2. Models 层(模型市场结构化数据) — `models/`
直接从百炼 console 网关 `listFoundationModels` 抓取的**结构化模型元数据**,
对应「模型广场」页面。涉及具体模型的能力、上下文长度、QPM 限流、定价、
官方 sample code 等问题时,优先查这里,比 wiki 更新更准。
**三层产物,单一数据源**(`families.jsonl` / `models.jsonl` 是机器查询入口,`index.md` 是 `families.jsonl` 的人类可读视图):
- `models/index.md` — 按主能力分桶的家族索引(中文标签 + 模型代码 + 链接到明细 JSON),适合「概览扫一眼」
- `models/families.jsonl` — **每行一个家族**,含 `slug` / `name` / `description` / `primaryCapability` /
`capabilities`(家族下所有 item 的并集)/ `providers` / `itemCount` / `maxContextWindow` / 轻量 `items[]`
摘要(每个 item 只含 `model` / `name` / `contextWindow` / `capabilities` / `openSource`)/ `detailPath`。
适合按家族筛选,例如:
- 列出推理类家族:`jq -c 'select(.primaryCapability=="Reasoning") | {slug,name,itemCount}' families.jsonl`
- 找包含某个 provider 的家族:`grep '"moonshot-ai"' families.jsonl | jq -c .slug`
- `models/models.jsonl` — **每行一个主干模型**(跨家族扁平),含 `model` / `family` / `capabilities` /
`features` / `contextWindow` / `maxInputTokens` / `maxOutputTokens` / `prices`(精简)/ `qpmInfo`(精简)/
`docUrl` / `detailPath` / **`profile`**(选型画像,见下)。适合跨家族批量查询,例如:
- 列出所有 contextWindow ≥ 1M 的模型:`jq -c 'select(.contextWindow>=1000000)' models.jsonl`
- 找支持 function-calling 的文本模型:`grep '"function-calling"' models.jsonl | jq -c '{model,family,contextWindow}'`
- 按场景选模型:`jq -c 'select(.profile.scenes // [] | index("代码生成")) | {model, pos: .profile.oneLinePositioning}' models.jsonl`
- 同族里挑旗舰:`jq -c 'select(.profile.familyTier=="flagship") | {model,family}' models.jsonl`
**`profile` —— LLM 合成的选型画像(判断层)**,用于回答「这个模型适合干什么、族里该选哪个」:
| 字段 | 含义 |
| --- | --- |
| `profileCompleteness` | `full`(判断字段齐备)/ `partial`(有定位和场景)/ `structured-only`(仅由结构化规格推导,未读过文档) |
| `oneLinePositioning` | 一句话定位(≤50 字) |
| `scenes` | 适用场景标签(受控词表,31 个值) |
| `notSuitableFor` | 不适用场景负面清单 |
| `styleKeywords` | 风格关键词(≤5 个) |
| `familyTier` | 族内选型档位:`flagship` 效果首选 / `balanced` 默认均衡 / `lightweight` 成本延迟优先或旧版 |
| `familyComparison` | 同族对比摘要(族内所有成员共享同一段文本) |
| `callModes` | 调用模式:`sync` / `async` / `streaming` / `realtime`(脚本从示例代码机械推导) |
| `pipelineRole` | 多模态流水线角色 `{from, to, role}` |
| `_meta` | `generatedAt` / `sourceHash` / `generatorVersion`,用于判断画像是否过期 |
**三态语义要读准**:字段**缺失** = 未知;`null` = 不适用(仅 `familyTier` / `familyComparison` 会出现,表示单模型家族);有值 = 已知。
`notSuitableFor: []` 意为「无已知限制」,与字段缺失(未知)不同。
`profileCompleteness` 为 `structured-only` 时该画像**没读过文档**,只有机械推导字段可信;模型的硬规格(价格 / QPM / contextWindow / `openSource`)永远以宿主记录的顶层字段为准,不要从画像里找。
`index.json` 的 `profileCoverage` 给出三档分布,可一眼判断整体画像新鲜度。
两份 JSONL 的 **join 字段**:`models.jsonl[].family == families.jsonl[].slug`。命中后按 `detailPath` 打开 `groups/<slug>.json` 取完整字段(`samples`、`predictConfig`)。
- `models/groups/<slug>.json` — 单个模型家族(如 `qwen3-max`、`deepseek`、`wan-image-to-video`)的完整明细:
- 家族层:`name`、`description`
- `items[]`:该家族下所有**主干模型版本**(剔除了带日期后缀的快照),保留字段:
`model`(API 调用名)、`modelAlias`、`contextWindow`、`maxInputTokens`、`maxOutputTokens`、
`capabilities`、`features`、`provider`、`docUrl`、`prices`、`qpmInfo`、
`samples`(**调用示例**,扁平化结构:`samples.<sdk>.<api>.{curl,python,nodejs,java,docUrl}`,
例如 `samples.openai.completionsAPI.python` 直接是代码字符串)、
`predictConfig`(**模型调用入参定义**,含 `system`/`temperature`/`top_p`/`enable_search`/`enable_thinking` 等参数的
`name`/`key`/`default`/`tip`/`range`,与百炼 Playground 对齐)
- `models/meta.json` — 增量爬取的指纹缓存,**不要给用户引用**
#### 能力代码(capability)映射
`index.md` 的章节标题与 `items[].capabilities[]` 用的是英文短代码,
对应中文含义如下:
| 代码 | 中文标签 |
| --- | --- |
| `TG` | 文本生成 |
| `Reasoning` | 推理 |
| `VU` | 视觉理解 |
| `IG` | 图像生成 |
| `VG` | 视频生成 |
| `TTS` | 语音合成 |
| `ASR` | 语音识别 |
| `Realtime-ASR` | 实时语音识别 |
| `Realtime-Text-to-Speech` | 实时语音合成 |
| `Realtime-Audio-Translate` | 实时音频翻译 |
| `Realtime-Omni` | 实时全模态 |
| `Multimodal-Omni` | 全模态 |
| `ME` | 多模态嵌入 |
| `TR` | 翻译 |
| `3D-generation` | 3D 生成 |
| `World-Model` | World-Model |
| `Realtime-Chatting` | Realtime-Chatting |From modelstudioai/skills, Apache-2.0.
More from modelstudioai/skills
- 0.630score
vox-video-director
modelstudioai
Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end with Aliyun Bailian CLI + local ffmpeg — script, collage keyframes…
SkillBFreshness: B★ 57Claude Code +9 - 0.630score
bailian-train-deploy
modelstudioai
用百炼 CLI (`bl`) 走完"数据→微调训练→导出→部署→调用"的完整闭环,或跳过训练直接部署基座模型。支持文本模型(SFT/DPO/CPT)、音频 TTS 模型(CosyVoice)、图像生成模型(Wan2.7)和视频生成模型(Wan…
SkillBFreshness: B★ 57Claude Code +9 - 0.625score
financial-expert
modelstudioai
Use when users need China or Hong Kong securities, funds, fund managers, company financials, valuation, global macro or industry time series, broker research…
SkillBFreshness: B★ 57Claude Code +9 - 0.577score
bailian-model-recommend
modelstudioai
阿里云百炼模型选型与推荐。当用户需要"选择、推荐、对比"模型,或描述一个 AI 场景/功能需求(隐含需要帮其决策用哪个模型)时激活。核心意图是帮用户做决策, 不是单纯查资料。触发词:推荐模型、选哪个、哪个适合、对比、做一个XX、实现XX 功能、用什么模型好、XX场景方案。当用户同时涉及模型查询和模型选择时,优先…
SkillCFreshness: C★ 57Claude Code +9 - 0.576score
spark-video-episode
modelstudioai
One-shot autopilot orchestrator — runs the full spark-video pipeline (screenwriter ↔ director per-scene parallel → render chain-DAG parallel + per-clip review…
SkillCFreshness: C★ 57Claude Code +9 - 0.575score
happyhorse-prompt-studio
modelstudioai
Interactive prompt studio for HappyHorse 1.0 video generation.
SkillCFreshness: C★ 57Claude Code +9 - 0.568score
novel-game
modelstudioai
将小说/故事改编为互动小说网页游戏(React SPA),含 AI 生成素材(视频/图片可选)、 可选 TTS 旁白、程序化音频、分支剧情引擎、存档系统。 使用 `bl` CLI 完成素材生成(`bl video generate` / `bl image generate` / `bl speech…
SkillCFreshness: C★ 57Claude Code +9
Signals
- Stars
- 57
- Forks
- 8
- Open issues
- 1
- Last commit (this folder)
- Oct 23, 2026 (5 days ago)
- Last release
- Jul 13, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- Apache-2.0
- Repository
- github.com/modelstudioai/skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from modelstudioai
- 0.630score
vox-video-director
modelstudioai
Turn ONE topic into a finished Vox-style paper-collage explainer / ad video, end to end with Aliyun Bailian CLI + local ffmpeg — script, collage keyframes…
SkillBFreshness: B★ 57Claude Code +9 - 0.630score
bailian-train-deploy
modelstudioai
用百炼 CLI (`bl`) 走完"数据→微调训练→导出→部署→调用"的完整闭环,或跳过训练直接部署基座模型。支持文本模型(SFT/DPO/CPT)、音频 TTS 模型(CosyVoice)、图像生成模型(Wan2.7)和视频生成模型(Wan…
SkillBFreshness: B★ 57Claude Code +9 - 0.625score
financial-expert
modelstudioai
Use when users need China or Hong Kong securities, funds, fund managers, company financials, valuation, global macro or industry time series, broker research…
SkillBFreshness: B★ 57Claude Code +9 - 0.577score
bailian-model-recommend
modelstudioai
阿里云百炼模型选型与推荐。当用户需要"选择、推荐、对比"模型,或描述一个 AI 场景/功能需求(隐含需要帮其决策用哪个模型)时激活。核心意图是帮用户做决策, 不是单纯查资料。触发词:推荐模型、选哪个、哪个适合、对比、做一个XX、实现XX 功能、用什么模型好、XX场景方案。当用户同时涉及模型查询和模型选择时,优先…
SkillCFreshness: C★ 57Claude Code +9 - 0.576score
spark-video-episode
modelstudioai
One-shot autopilot orchestrator — runs the full spark-video pipeline (screenwriter ↔ director per-scene parallel → render chain-DAG parallel + per-clip review…
SkillCFreshness: C★ 57Claude Code +9 - 0.575score
happyhorse-prompt-studio
modelstudioai
Interactive prompt studio for HappyHorse 1.0 video generation.
SkillCFreshness: C★ 57Claude Code +9
Cite this page
bailian-docs-llm-wiki skill. Agent Mart. https://agentmart.dev/skills/modelstudioai-bailian-docs-llm-wiki (last updated 2026-10-24)