Agno website screenshot

Agno

Agno (formerly Phidata) is a high-performance open-source Python framework for building multi-modal, multi-agent systems with memory, knowledge, tools, and reasoning capabilities. It provides both an agent development SDK and AgentOS, a production runtime that transforms agents into managed APIs with 50+ endpoints for runs, sessions, memory, knowledge, traces, evaluations, and scheduling. AgentOS supports RBAC, per-user session isolation, streaming via Server-Sent Events, background execution, and human-in-the-loop approval flows. The platform supports 30+ LLM providers and can be deployed on AWS, GCP, Railway, or air-gapped environments with data remaining in the customer's cloud.

1 APIs 0 Features
AI AgentsMulti-Agent SystemsPython FrameworkLLMMachine LearningAgentic AIOpen SourceAgent Runtime

APIs

AgentOS Agent API

The AgentOS Agent API provides 50+ RESTful endpoints for managing agents in production. Endpoints cover runs (create, list, cancel, resume), sessions (user-scoped with full hist...

Pricing Plans

Agno Plans Pricing

3 plans

PLANS

Rate Limits

Agno Rate Limits

3 limits

RATE LIMITS

FinOps

Agno Finops

FINOPS

Semantic Vocabularies

Agno Context

14 classes · 37 properties

JSON-LD

Example Payloads

Agno Agentos Api Examples

4 fields

EXAMPLE

Resources

🔗
Website
Website
🔗
Documentation
Documentation
👥
GitHubOrganization
GitHubOrganization
🔗
LinkedIn
LinkedIn
🔗
X
X
📰
Blog
Blog
📄
ChangeLog
ChangeLog
💰
Pricing
Pricing
🔗
Plans
Plans
🔗
RateLimits
RateLimits
🔗
FinOps
FinOps
🔗
Vocabulary
Vocabulary
🔗
JSONLDContext
JSONLDContext
💻
Examples
Examples

Sources

Raw ↑
aid: agno
name: Agno
description: Agno (formerly Phidata) is a high-performance open-source Python framework for building multi-modal, multi-agent
  systems with memory, knowledge, tools, and reasoning capabilities. It provides both an agent development SDK and AgentOS,
  a production runtime that transforms agents into managed APIs with 50+ endpoints for runs, sessions, memory, knowledge,
  traces, evaluations, and scheduling. AgentOS supports RBAC, per-user session isolation, streaming via Server-Sent Events,
  background execution, and human-in-the-loop approval flows. The platform supports 30+ LLM providers and can be deployed
  on AWS, GCP, Railway, or air-gapped environments with data remaining in the customer's cloud.
type: Index
image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
tags:
- AI Agents
- Multi-Agent Systems
- Python Framework
- LLM
- Machine Learning
- Agentic AI
- Open Source
- Agent Runtime
url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/apis.yml
created: '2026-06-12'
modified: '2026-06-12'
specificationVersion: '0.19'
apis:
- aid: agno:agentos-api
  name: AgentOS Agent API
  description: The AgentOS Agent API provides 50+ RESTful endpoints for managing agents in production. Endpoints cover runs
    (create, list, cancel, resume), sessions (user-scoped with full history), memory (CRUD with search and filtering), knowledge
    (file and URL uploads with vector/keyword/hybrid search), evaluations, traces (full span trees with filtering DSL), metrics
    (daily aggregated usage and token analytics), schedules (recurring agent tasks), and approvals (human-in-the-loop gating).
    Authentication uses JWT Bearer tokens with role-based access control. The API is FastAPI-based with live docs at /docs
    and an OpenAPI spec at /openapi.json. Multipart form-data supports combined text, files, and media in a single call.
  image: https://kinlane-images.s3.amazonaws.com/shared/apis-json/apis-json-logo.jpg
  humanURL: https://docs.agno.com/features/api
  baseURL: https://api.agno.com
  tags:
  - AI Agents
  - Agent Runtime
  - Multi-Agent
  - Sessions
  - Memory
  - Knowledge
  properties:
  - type: Documentation
    url: https://docs.agno.com/features/api
  - type: OpenAPI
    url: https://docs.agno.com/openapi.json
  - type: OpenAPI
    url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/openapi/agno-agentos-api-openapi.yml
common:
- type: Website
  url: https://www.agno.com/
- type: Documentation
  url: https://docs.agno.com/
- type: GitHubOrganization
  url: https://github.com/agno-agi
- type: LinkedIn
  url: https://www.linkedin.com/company/agno-agi
- type: X
  url: https://x.com/AgnoAgi
- type: Blog
  url: https://www.agno.com/blog
- type: ChangeLog
  url: https://www.agno.com/changelog
- type: Pricing
  url: https://www.agno.com/pricing
- type: Plans
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/plans/agno-plans-pricing.yml
- type: RateLimits
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/rate-limits/agno-rate-limits.yml
- type: FinOps
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/finops/agno-finops.yml
- type: Vocabulary
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/vocabulary/agno-vocabulary.yml
- type: JSONLDContext
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/json-ld/agno-context.jsonld
- type: Examples
  url: https://raw.githubusercontent.com/api-evangelist/agno/refs/heads/main/examples/agno-agentos-api-examples.json
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com