Synthetic Sciences 发布了 OpenScience，这是一个面向科学研究的开源 AI 工作台。它采用 Apache 2.0 许可证，运行在你自己的基础设施上。研究团队将其定位为 Anthropic 于 2026 年 6 月底推出的 Claude Science 的开源替代方案。
其主张很直接。科学领域的 AI 工具不应由某一家供应商独占。OpenScience 保持工作流程开放、模型可替换、数据本地化。这是一个独立项目，与 Anthropic 无关，也未获得其认可。
摘要
- OpenScience 是一个采用 Apache-2.0 许可证、与模型无关的 AI 工作台，适用于机器学习、生物学、物理学和化学领域。
- 它运行完整的流程：文献、假设、代码、实验、分析和撰写报告。
- 任何模型都能使用（Claude、GPT、Gemini、GLM、Kimi、DeepSeek、本地微调模型）；每次请求均可切换模型。
- 它内置了 250 多个可编辑的技能，以及数据库（UniProt、PDB、ChEMBL、arXiv 以及约 30 个其他数据库）作为智能体工具。
- 它在你自己的基础设施上运行，使用你自己的密钥；自带密钥的使用是免费的，且永不受限。
什么是 OpenScience
OpenScience 是一个基于浏览器的交互空间，由本地智能体运行时支持。你给它一个研究目标，它就会按照一个称职的合作者会遵循的流程来工作。
它阅读相关论文，形成假设，编写并运行代码，执行实验。它查询主要的科学数据库，并撰写结果报告。所有这些都在一个连续的会话中完成。
该工具在设计上与模型无关。它可以使用任何前沿或开源权重模型，并使用你自己的 API 密钥。无需注册账号即可开始使用。
安装使用 npm。命令是 openscience，它会在你的浏览器中打开工作台。
npm install -g @synsci/openscience
openscience首次运行提供三个选项：Atlas 托管模型、你自己的提供商密钥，或免费演示模型。你也可以跳过全局安装。运行 npx synsci 可以一步完成相同操作。
工作原理
OpenScience 运行一个本地服务器。该服务器托管工作台 UI、智能体运行时和工具层。智能体使用一个研究框架进行规划并调用工具。
这些工具包括 shell、编辑器、LSP、MCP 服务器、科学连接器和技能。智能体在运行时将其工作流实时传输回浏览器。
模型按请求进行路由。你可以从工作区的模型选择器中选取模型。因此，你可以在不更改任何其他设置的情况下切换提供商或运行本地模型。
# Bring your own key; requests go straight to the provider
export ANTHROPIC_API_KEY=sk-ant-...
openscience
# Or open a specific project directory
openscience ~/code/my-project你的密钥保留在你的机器上。会话、工件和来源信息存储在磁盘上。它们可以作为链接共享。
有四个要素使该运行时对实际工作有用：
- 研究智能体：默认运行一个研究智能体。此外，还存在生物学、物理学和机器学习领域的专业智能体。还包括批评和文献综述子智能体，以及一个只读的计划模式。
- 250 多项技能：这些技能涵盖训练（DeepSpeed、PEFT、TRL）、评估、数据集工作和化学信息学。它们还涵盖分子与临床生物学、论文、LaTeX、图表和云计算。
- 作为工具的科学数据库：UniProt、PDB、Ensembl、ChEMBL、PubChem、arXiv、OpenAlex 和 Semantic Scholar 均可查询。还包括大约 30 个其他数据库。
- 一个真正的工作区：它拥有文件树、编辑器、终端和会话历史。它能够内联渲染分子、结构、基因组和图表。
可扩展性是一等特性。OpenScience 支持 LSP 集成、MCP 服务器、插件和自定义智能体。它还附带了一个 TypeScript SDK。
还有一个可选的管理层，名为 Atlas。Atlas 提供一组精选的前沿模型，通过预付费钱包计费。它还增加了一个持久的研究图谱和云计算能力。OpenScience 可与 Atlas 配合使用，但从不强制要求。
OpenScience 对比 Claude Science
这两个工具针对的是相同的工作任务。两者都运行循环，内联渲染科学内容，并优先考虑可复现性。核心区别在于开放性和模型选择。
| 维度 | OpenScience | Claude Science |
|---|---|---|
| 供应商 | Synthetic Sciences | Anthropic |
| 许可证 | 开源，Apache 2.0 | 专有产品 |
| 模型 | 任何提供商或本地微调模型 | 仅限 Anthropic Claude 模型 |
| 模型切换 | 按请求，通过模型选择器 | 固定为 Claude |
| 密钥 / 成本 | 使用你自己的密钥；自带密钥免费，永不设限 | 需要付费的 Claude 订阅 |
| 技能 / 工具 | 250 多项可编辑、可扩展的技能 | 60 多项精选技能和连接器 |
| 运行环境 | 你的基础设施，浏览器工作区 | 实验室机器；macOS 和 Linux 上为测试版 |
| 子智能体 | 研究、生物学、物理学、机器学习 + 批评 | 协调智能体 + 专家 + 评审员 |
| 数据库 | UniProt、PDB、ChEMBL、arXiv，以及约30个其他数据库 | UniProt、PDB、ChEMBL、GEO 等 |
| 专用模型 | 使用你选择的任意模型 | 接入 NVIDIA BioNeMo（Evo 2、Boltz-2、OpenFold3） |
Claude Science 是一个打磨精良、自带精选集成的独立产品。OpenScience 则在开放性、可审计性和供应商自由度方面做出了一些取舍，以换取更少的打磨。
用例示例
- 机器学习研究：一位机器学习工程师想测试一个微调思路。ML 智能体拉取相关的 arXiv 论文，然后使用 PEFT 和 TRL 技能。它编写训练脚本、运行脚本，并草拟一份简短报告。
- 计算生物学：一位数据科学家研究某个蛋白质靶点。生物学智能体查询 UniProt 和 PDB，然后内联渲染结构。它提出候选突变方案并记录来源。
- 化学信息学：一位化学家筛选小分子。智能体查询 ChEMBL 和 PubChem 获取生物活性数据。它在代码中运行过滤器，并返回带有图表的排序候选结果。
- 预算内的模型对比：一个团队在 Claude、GLM 以及本地微调模型上运行相同任务。切换模型只需一次选择，无需重写代码。他们用自己的数据比较成本和输出质量。
优势与不足
优势：
- 在 Apache 2.0 协议下完全开源，因此技能和智能体均可读可编辑。
- 模型无关的路由机制消除了科学工作流中对单一供应商的锁定。
- 在你的基础设施上运行，因此私有数据集可以保留在你的系统内。
- 广泛的工具覆盖：250 多项技能和数十个科学数据库作为工具。
- 可通过 LSP、MCP 服务器、插件和 TypeScript SDK 进行扩展。
不足：
- 智能体未进行沙箱隔离；权限系统并非隔离边界。
- 如果需要隔离，你应该在容器或虚拟机内运行它。
- 这是一个年轻的项目，因此与成熟产品相比，难免存在粗糙之处。
- 自带密钥意味着你需要自行管理供应商成本和速率限制。
- 输出质量在很大程度上取决于你将每个请求路由到哪个模型。
交互式说明
Synthetic Sciences has released OpenScience, an open-source AI workbench for scientific research. It is licensed under Apache 2.0 and runs on your own infrastructure. The research team frames it as an open alternative to Anthropic’s Claude Science, launched in late June 2026.
The pitch is direct. Scientific AI tooling should not be owned by one vendor. OpenScience keeps the workflow open, the models swappable, and the data local. It is an independent project, not affiliated with or endorsed by Anthropic.
TL;DR
- OpenScience is an Apache-2.0, model-agnostic AI workbench for machine learning, biology, physics, and chemistry.
- It runs the full loop: literature, hypothesis, code, experiment, analysis, and write-up.
- Any model works (Claude, GPT, Gemini, GLM, Kimi, DeepSeek, local fine-tunes); switching is per-request.
- It ships 250+ editable skills, plus databases (UniProt, PDB, ChEMBL, arXiv, and ~30 more) as agent tools.
- It runs on your infrastructure with your keys; bring-your-own-key usage is free and never gated.
What is OpenScience
OpenScience is a browser-based workspace backed by a local agent runtime. You give it a research goal. It then works through the loop a capable collaborator would follow.
It reads relevant papers, forms a hypothesis, writes and runs code, and runs experiments. It queries major scientific databases and writes up the result. All of this happens in one continuous session.
The tool is model-agnostic by design. It works with any frontier or open-weight model, using your own API keys. No account is required to start.
Installation uses npm. The command is openscience, and it opens the workspace in your browser.
npm install -g @synsci/openscience
openscienceThe first run offers three options: Atlas managed models, your own provider keys, or free demo models. You can also skip a global install. Running npx synsci does the same thing in one step.
How It Works
OpenScience runs a local server. That server hosts the workspace UI, the agent runtime, and the tool layer. The agent plans with a research harness and calls tools.
Those tools include the shell, editor, LSP, MCP servers, scientific connectors, and skills. The agent streams its work back to the browser as it runs.
Models are routed per request. You pick the model from the model selector in the workspace. So you can switch providers or run local models without changing anything else.
# Bring your own key; requests go straight to the provider
export ANTHROPIC_API_KEY=sk-ant-...
openscience
# Or open a specific project directory
openscience ~/code/my-projectYour keys stay on your machine. Sessions, artifacts, and provenance are stored on disk. They can be shared as links.
Four things make the runtime useful for real work:
- Research agents: A
researchagent runs by default. Specialistbiology,physics, andmlagents exist too. Critique and literature-review sub-agents and a read-only plan mode round it out. - 250+ skills: These cover training (DeepSpeed, PEFT, TRL), evaluation, dataset work, and cheminformatics. They also cover molecular and clinical biology, papers, LaTeX, figures, and cloud compute.
- Scientific databases as tools: UniProt, PDB, Ensembl, ChEMBL, PubChem, arXiv, OpenAlex, and Semantic Scholar are queryable. Around 30 more are included.
- A real workspace: It has a file tree, editor, terminal, and session history. It renders molecules, structures, genomes, and plots inline.
Extensibility is a first-class feature. OpenScience supports LSP integration, MCP servers, plugins, and custom agents. It also ships a TypeScript SDK.
There is an optional managed layer called Atlas. Atlas gives a curated set of frontier models billed from a prepaid wallet. It also adds a persistent research graph and cloud compute. OpenScience works with Atlas but never requires it.
OpenScience vs Claude Science
Both tools target the same job. Both run the loop, render science inline, and prioritize reproducibility. The core difference is openness and model choice.
| Dimension | OpenScience | Claude Science |
|---|---|---|
| Vendor | Synthetic Sciences | Anthropic |
| License | Open source, Apache 2.0 | Proprietary product |
| Models | Any provider or local fine-tune | Anthropic Claude models only |
| Model switching | Per-request, via model selector | Fixed to Claude |
| Keys / cost | Your keys; BYOK free, never gated | Paid Claude subscription required |
| Skills / tools | 250+ editable, extensible skills | 60+ curated skills and connectors |
| Where it runs | Your infrastructure, browser workspace | Lab machines; beta on macOS and Linux |
| Sub-agents | research, biology, physics, ml + critique | Coordinating agent + specialists + reviewer |
| Databases | UniProt, PDB, ChEMBL, arXiv, ~30 more | UniProt, PDB, ChEMBL, GEO, and others |
| Special models | Uses whatever model you pick | Taps NVIDIA BioNeMo (Evo 2, Boltz-2, OpenFold3) |
Claude Science is a polished, standalone product with curated integrations. OpenScience trades some polish for openness, auditability, and provider freedom.
Use Cases With Examples
- Machine learning research: An ML engineer wants to test a fine-tuning idea. The
mlagent pulls related arXiv papers, then uses PEFT and TRL skills. It writes a training script, runs it, and drafts a short report. - Computational biology: A data scientist studies a protein target. The
biologyagent queries UniProt and PDB, then renders the structure inline. It proposes candidate mutations and logs the provenance. - Cheminformatics: A chemist screens small molecules. The agent queries ChEMBL and PubChem for bioactivity data. It runs a filter in code and returns ranked candidates with plots.
- Model comparison on a budget: A team runs the same task on Claude, then GLM, then a local fine-tune. Switching is one selection, not a rewrite. They compare cost and quality on their own data.
Strengths and Weaknesses
Strengths:
- Fully open source under Apache 2.0, so skills and agents are readable and editable.
- Model-agnostic routing removes single-vendor lock-in for scientific workflows.
- Runs on your infrastructure, so private datasets can stay on your systems.
- Broad tool coverage: 250+ skills and dozens of scientific databases as tools.
- Extensible through LSP, MCP servers, plugins, and a TypeScript SDK.
Weaknesses:
- The agent is not sandboxed; the permission system is not an isolation boundary.
- You should run it inside a container or VM if you need isolation.
- It is a young project, so expect rough edges versus a mature product.
- Bring-your-own-key means you manage provider costs and rate limits yourself.
- Quality depends heavily on which model you route each request to.