smolagents (Hugging Face)

https://github.com/huggingface/smolagents

Hugging Face's minimal agent library — emphasises code-execution agents (agent writes Python, runs it) over JSON tool-calling. Library is intentionally ~1k LOC.

Active · Verified 2026-10-31 · fresh as of 2026-10-31 · 23 cited facts, 2 with a third-party source · 1 recorded change · Suggest a correction

At a glance

Tier
T2
Created
2024 repo fix
Latest release
v1.26.0 (2026-05-29) registry fix
License
Apache-2.0 repo fix
Pricing
searched not found fix
Funding
searched not found fix
Freshness
2026-10-31 fix
Layer
agent-infra repo fix
Licensing
open-source repo fix
Deployment locus
local repo fix
Platform vs product
platform repo fix
Accelerator
betaworks-camp-2016 third-party fix
SDK languages
python repo fix
Protocols
mcp-client repo fix

When to use

Optimised for: searched not found

Anti-fit: searched not found

Pros & cons

Pros

searched not found

Cons

searched not found

Claims & capabilities

Minimal HF agents library released 2024-12; <1k LOC; built-in tool-calling and code agents.

Technical surface

API surface
searched not found
Backend storage
not applicable — not a memory product
Deployment
Self-hosted only — library
Embedding model
not applicable — not a memory product
Multi-tenancy
searched not found
MCP
yes — MCPClient and ToolCollection.from_mcp load tools from stdio or Streamable HTTP MCP servers
A2A
searched not found
OpenTelemetry
searched not found

Similar systems

Other agent frameworks (no first-party memory layer) in the catalog, ranked by inbound references.

  • Model Context Protocol (MCP spec) T1

    Anthropic-introduced open protocol (Nov 2024) for connecting LLMs to data sources and tools. Donated by Anthropic to the Linux Foundation's Agentic AI Foundation (announced 2025-12-09); now governed as Model Context Protocol, a Series of LF Projects, LLC. Supported across Claude, ChatGPT, VS Code, Cursor and many other clients.

  • LangChain (framework) T1

    Dominant Python/JS framework for LLM apps — tool calling, agents, retrievers, runnables. Memory is one component (LangMem cross-listed); the framework itself is the product.

  • Vercel AI SDK T1

    Vercel's TypeScript LLM SDK — streaming, tool calls, structured outputs, agents. Dominant framework for Next.js LLM apps.

  • LlamaIndex (framework) T1

    Data framework for LLM apps — connectors, indices, query engines, agents, workflows. Memory is one of many subsystems (LlamaIndex Memory cross-listed).

  • OpenAI Agents SDK T1

    OpenAI's official Python framework for agents — handoffs, guardrails, tracing built in. Successor to OpenAI Swarm (experimental); officially supported.

  • LangGraph (orchestration) T1

    LangChain's stateful graph runtime for agents — explicit state, persistence, time-travel debugging, human-in-the-loop. Memory is one persistence target (LangGraph Persistence cross-listed).

Sources (22 cited facts, 2 third-party)

Each fact links to the page it came from; the badge says who published it (how to read sources).

What changed

Changes to this record's lifecycle, licence, release, pricing and hosting facts, from the catalog's git history. Subscribe (Atom)

  1. 2026-10-27
    • Latest release first recorded as “v1.26.0 (2026-05-29)” source
    • Deployment first recorded as “Self-hosted only — library” source
    • MCP support first recorded as “yes — MCPClient and ToolCollection.from_mcp load tools from stdio or Streamable HTTP MCP servers” source
    #270

Related systems

References (10)

  • Arize Phoenix integrates with — smolagents inspect-runs tutorial has a section on setting up telemetry with Arize Phoenix. (verified 2026-10-29)
  • e2b integrates with — smolagents secure code execution docs have an E2B sandbox setup section. (verified 2026-10-29)
  • LangChain (framework) integrates with — smolagents tools tutorial covers using LangChain tools. (verified 2026-10-29)
  • Langfuse integrates with — smolagents inspect-runs tutorial has a section on setting up telemetry with Langfuse. (verified 2026-10-29)
  • LiteLLM integrates with — smolagents models reference documents LiteLLMModel and LiteLLMRouterModel. (verified 2026-10-29)
  • MLflow integrates with — smolagents inspect-runs tutorial has a section on setting up telemetry with MLflow. (verified 2026-10-29)
  • Modal Sandboxes integrates with — smolagents secure code execution docs have a Modal sandbox setup section. (verified 2026-10-29)
  • Model Context Protocol (MCP spec) integrates with — smolagents tools tutorial covers using tools from an MCP server. (verified 2026-10-29)
  • OpenAI GPT family (GPT-5 / GPT-4o / o3 / o4) integrates with — smolagents models reference documents OpenAIModel. (verified 2026-10-29)
  • vLLM integrates with — smolagents models reference documents VLLMModel. (verified 2026-10-29)

Referenced by (8)

  • AgentOps integrates with — AgentOps docs have a smolagents (Hugging Face) integration page. (verified 2026-10-29)
  • Arize Phoenix integrates with — Arize Phoenix docs have a smolagents (Hugging Face) integration page. (verified 2026-10-29)
  • Comet ML integrates with — Comet's Opik docs have a smolagents (Hugging Face) integration page. (verified 2026-10-29)
  • Langfuse integrates with — Langfuse's integrations directory has a smolagents (Hugging Face) integration page. (verified 2026-10-29)
  • OpenInference integrates with — The OpenInference repo README lists an OpenInference instrumentation package for smolagents (Hugging Face). (verified 2026-10-29)
  • Pydantic Logfire integrates with — Pydantic Logfire integrations docs have a smolagents page. (verified 2026-10-29)
  • Qdrant integrates with — Qdrant's documentation has a smolagents (Hugging Face) integration page. (verified 2026-10-29)
  • Weights & Biases integrates with — W&B Weave docs have a smolagents (Hugging Face) integration page. (verified 2026-10-29)