How does AgentScope use Jev?

It ships JevClassifierModel, which calls the TypeSafe SDK to answer binary, choice and score classification questions, so routing decisions can run through Jev instead of a generative model.

Is AgentScope tied to Alibaba infrastructure?

No. It is developed by Alibaba and documented first with DashScope, but it supports OpenAI, Anthropic, Gemini, DeepSeek, Moonshot, xAI, Ollama and other chat model providers.

How do I install AgentScope, and what does it require?

Python 3.11 or higher. From PyPI it is `uv pip install agentscope`, or clone the repository and run `uv pip install -e .` for an editable source install. The repository is Apache-2.0 licensed and documents itself at docs.agentscope.io in English and Chinese.

What else is in the SDK besides the Jev classifier model?

Agent building blocks: a ReAct agent with structured output and batched tool acting, a Toolkit over Python tools, MCP servers and skills with built-in shell, file edit and search tools, a Model layer across providers, Context middleware that compacts and offloads tool results, an event bus, permission and human-in-the-loop control, memory on ReMe or Mem0, and workspaces from local and Docker to Bubblewrap, E2B, Daytona and Kubernetes.

Do I have to build a backend and web UI myself?

No. The project ships a batteries-included agent service: a FastAPI backend with a prebuilt web UI under examples/web_ui, plus multi-tenancy and session isolation. For terminal work, launch_console(agent) handles streamed output, tool-call confirmation and Ctrl+C interruption.

What shipped alongside the Jev integration?

Several features landed in the same release: ModelRouterMiddleware selects an appropriate chat model per reply, TeamPipeline lets a leader agent delegate to member agents, SOP runs multi-step tasks as standard operating procedures, A2AAgent speaks the A2A protocol with remote agents, and realtime voice works over WebRTC with text and voice sharing one session history.

Which parts are still marked experimental?

The release notes tag realtime voice conversations over WebRTC, the SOP feature and the realtime voice agent as Experimental. ModelRouterMiddleware, TeamPipeline, the A2A protocol and the Jev classifier integration are listed as regular features or integrations rather than experimental work.

CommunityJev Integrations & Adapters

AgentScope

Alibaba's multi-agent platform, which implements native TypeSafe Jev classification through JevClassifierModel for binary, choice and score routing across agent pipelines. Python 3.11+, Apache-2.0, with a FastAPI agent service and web UI.

See highlights

Open github.com

Stars: 33k
Language: Python
License: APACHE-2.0
Published by: Community
Last verified: Oct 2026

agentscope-ai/agentscope

Organization repository on GitHub

View on GitHub

Build and run agents you can see, understand and trust.

Forks: 3.6k
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

About this project

AgentScope 2.0 is a production-ready, easy-to-use agent framework with essential abstractions that keep up with rising model capability. We design for increasingly agentic LLMs. Our approach leverages the models' reasoning and tool use abilities rather than constraining them with strict prompts and opinionated orchestrations. [2026-10] FEAT Experimental: Agent service supports realtime voice conversations over WebRTC, with text and voice sharing the same session history.

Quoted from the project's README. Read the full README.

Tagscommunity multi-agent framework alibaba

Works forJev for Routing Jev for Classification Jev for Scoring

Used inJev for Software & DevTools

Languages: Python
Platforms: API MCP
Capabilities: Open sourceSelf-hostedRequires API key

Highlights

September 2026 news: Jev classifier model supported through the TypeSafe SDK via JevClassifierModel.
The same release added ModelRouterMiddleware for selecting an appropriate chat model per reply.
Ships a batteries-included agent service: FastAPI backend, prebuilt web UI, multi-tenancy and session isolation.
Roughly 33k stars; Apache-2.0; requires Python 3.11 or higher and supports OpenAI, Anthropic, Gemini, DashScope and Ollama models.

Watch out

A broad framework where Jev is one supported model among many; the Jev path needs a TypeSafe key. Apache-2.0 licensed.

Questions about this entry

Short answers drawn from the linked project or page.

More Jev resources like this

Jev-Mail

CLI-installed Gmail classifier that runs 24/7 on user-owned Google Apps Script: the worker asks Jev for three typed judgments per thread (requires_action, is_important, bucket), applies labels, stars, and optional archiving, and routes ambiguous or low-confidence mail to a Review label.

Jev Integrations & Adapters#community#gmail#email

Community

vynnlee/jev-mailTypeScript

AutoGPT

187.5k

Autonomous agent platform with 187,504 stars for building and running agents from plain English or a visual builder. Its platform features first-class TypeSafe Jev decision blocks for typed routing, filtering, scoring and confidence-gated next-action dispatching.

Jev Integrations & Adapters#community#autonomous-agents#platform

Community

Significant-Gravitas/AutoGPTPython

LiteLLM

60.9k

LiteLLM's proxy adds TypeSafe guardrail hooks and a Jev complexity classifier for policy judgments and request routing across 100+ model providers.

Jev Integrations & Adapters#community#python#gateway

Community

BerriAI/litellmPython

Eliza

19.6k

Open-source TypeScript agentic operating system: a core runtime, app for web and mobile, CLI, cloud services and first-party plugins. The directory's curated summary reports it integrates TypeSafe System One decision services for sub-100ms intent classification, action dispatching and confidence-gated tool execution.

Jev Integrations & Adapters#community#agents#framework

Community

elizaOS/elizaTypeScript

JarvisCore

192

Python multi-agent runtime that ships Jev natively in an optional extras install: agents ask typed Choice, Score and Noul questions through a shared decision client, the Kernel can pick specialist subagents by Choice, and RAG passages with a prompt-injection Noul above 0.70 are withheld.

Jev Integrations & Adapters#community#framework#mesh

Community

Prescott-Data/jarviscore-frameworkPython

WorldMonitor

88.2k

Real-time geopolitical intelligence dashboard that uses TypeSafe Jev questions to score news headline severity into five threat levels and categorize events across 14 conflict, cyber and infrastructure domains, alongside news aggregation and a Country Instability Index.

Jev Integrations & Adapters#community#osint#geopolitics

Community

koala73/worldmonitorTypeScript

Back to all resources

From the Jev community

Posts from builders shipping with Jev right now.

Follow @typesafeai