# DP Technology (Bohrium)

**Canonical:** https://apis.io/providers/dp/  
**Website:** https://dp.tech  
**APIs profiled:** 15

DP Technology (Deep Potential Technology, 深势科技) is a global pioneer of the "AI for Science" research paradigm, founded in 2018 with offices in Beijing, Shanghai, Shenzhen, and Yibin. Its Bohrium platform is a research cloud for microscale scientific computing and industrial design — compute jobs, development nodes, datasets, container images, projects, knowledge bases, large knowledge models, paper/patent/scholar search, PDF parsing, a scientific-tools library, and web search — all reachable through the Bohrium OpenAPI (Bearer AccessKey auth, base URL https://open.bohrium.com). DP Technology also founds and maintains the open-source DeepModeling molecular-simulation toolchain (DeePMD-kit, DP-GEN, dpdispatcher, dpdata) and ships a first-party Python SDK, agent SDK, CLI, and 17 official Agent Skills.

## Kin Score — 35.4 / 100 (thin)

Scored 2026-08-20 under rubric 0.12.0. Trend: flat (+0.0 from 35.4).

| Facet | Score |
|---|---|
| Discoverability | 81.5 |
| Contract Quality | 45.2 |
| Governance | 16.7 |
| Contract Governance | 16.7 |
| Operational Transparency | 2.6 |
| Developer Ergonomics | 54.8 |
| Commercial Clarity | 13.2 |
| Access Clarity | 13.2 |

## Agent readiness — 33.8 (agent-aware)

| Dimension | Value |
|---|---|
| Spec Presence | yes |
| Agentic Access | derived |
| Reversibility Documented | no |
| MCP Server | no |
| Auth Clarity | yes |
| Idempotency | no |
| Error Semantics | documented |
| OpenAPI Examples | no |
| Rate Limit Signal | no |
| Event Surface Described | no |
| Agent Skills | yes |
| Well Known Catalog | no |
| Consent Identity | no |
| Agent Card | no |
| Dry Run Mode | no |

## Access

Self-serve signup — onboarding: self-serve, pricing: unknown, trial: no (confidence: medium).

## APIs (15)

- **DP Technology (Bohrium) AI 科学小导师 (bohrium-mentor) API** — 基于深度推理的科学问答（sigma-search，SSE 编排）。收费：创建会话 2 元/次。注意内部版本混用（sessions v4 / SSE v3 / history v4）。
- **DP Technology (Bohrium) 数据集 (bohrium-dataset) API** — 数据集管理 — 创建、上传、下载、版本控制。免费。
- **DP Technology (Bohrium) 文件盘 (bohrium-file) API** — personal/share 盘文件管理 — 列出、上传、下载、移动、复制、删除。v1 用于 personal/share，v2 file 仅用于 appJob。免费。
- **DP Technology (Bohrium) 容器镜像 (bohrium-image) API** — 容器镜像管理 — 查询、拉取、创建、删除镜像。免费。镜像地址位于 registry.dp.tech / registry.bohrium.dp.tech。
- **DP Technology (Bohrium) 计算任务 (bohrium-job) API** — 计算任务管理 — 提交、查询、终止、删除任务（提交主要经 bohr CLI）。任务运行按机时收费，价格见 Job 定价页。
- **DP Technology (Bohrium) 知识库 (bohrium-knowledge-base) API** — 知识库管理 — 知识库/文件夹/文献/标签/笔记/召回/权限。网关转发至 literature-sage。免费（仅容量配额）。
- **DP Technology (Bohrium) 大知识模型 (bohrium-lkm) API** — LKM — 知识节点检索、推理链检索、论文知识图谱、追溯 claim 依据、批量节点水合、提交反馈。收费（定价中，待补充）。
- **DP Technology (Bohrium) 开发节点 (bohrium-node) API** — 开发节点管理 — 创建、启停、删除容器/虚拟机。运行按机时收费，价格见 Node 定价页。
- **DP Technology (Bohrium) 论文与专利搜索 (bohrium-paper-search) API** — RAG 引擎关键词+语义检索。收费。响应可能为多行 JSON（流式，解析第一行）。
- **DP Technology (Bohrium) 项目管理 (bohrium-project) API** — 项目管理 — 创建项目、管理成员、设置额度。免费（管理成本额度，非按调用计费）。
- **DP Technology (Bohrium) 学者搜索 (bohrium-scholar-search) API** — 学者搜索与画像 — 按姓名/机构检索，查看发文/引用/h-index/研究方向。免费。
- **DP Technology (Bohrium) 科学百科 (bohrium-sciencepedia) API** — 科学百科 — 搜词条/关键词、浏览领域课程、看章节知识点、查主题知识图谱。免费（限时）。阅读链接指向 www.bohrium.com。
- **DP Technology (Bohrium) 科学工具库 (bohrium-tools) API** — 科学工具库 — 按领域/子领域浏览、混合检索工具、查看详情与分类。免费。语言经 Content-Language 头或 body language 指定。
- **DP Technology (Bohrium) 网页搜索 (bohrium-web-search) API** — 网页搜索 — 代理 searchapi.io 做开放互联网检索。免费。
- **DP Technology (Bohrium) PDF 解析 (bohrium-pdf-parser) API** — Uni-Parser — 提取文本、表格、图表、公式。收费：0.05 元/页（触发时扣）。figure 模块无权限时可能 403。

## MCP servers (1)

- **DP Technology (Bohrium) MCP Server**

## Agentic access (1)

- **Dp Agentic Access** — 166 operations · 99 acting

## Security (2)

- **Dp Authentication** — http · 1 scheme
- **Dp Domain Security** — TLSv1.2 · HSTS · DMARC

## Tags

Company, AI for Science, Scientific Computing, Molecular Simulation, Drug Discovery, Materials Science, Research Cloud, Machine-Learning, HPC

---

Profiled by [API Evangelist](https://apievangelist.com) and published on [APIs.io](https://apis.io/providers/dp/). Scores are computed from the provider's own public artifacts under a published rubric.