Agentic WeeklyGet the weekly
567 Labsv1.17.0Tested (smoke test)

Instructor: what’s new in v1.17.0 and our verdict

Python library for getting validated, structured outputs from language models using Pydantic models.

Official site

Release Ruler for Instructor: marked at its Tested release date on the timeline of every AI model we track.

At a glance

Evidence
Tested (smoke test)
Released
Sep 9, 2026
Last verified
Oct 25, 2026
Pricing
Partly verified
Best for
agent building

Evidence and ranking

Evidence behind this verdict

  1. Sources read2 cited sources, latest fetched Oct 25, 2026. (done)
  2. Facts re-checkedLast verified Oct 25, 2026. (done)
  3. Run by usSmoke test on Oct 25, 2026 (scope below). (done)

Passed Test log result: Passed. Read the log

Where we rank it (editorial)

Listed here but not yet ranked. Ranks are editorial picks, not benchmark scores.

Pricing: Open source; no pricing page read. (checked 2026-10-25)

Our verdict

A small, light way to get typed output from a model; good first step before reaching for a full framework.

Good for

  • Developers who need typed, validated model output
  • Teams already using Pydantic

Skip if

  • You need an agent runtime with tools and memory

Tested. Tested 2026-10-25 (America/Chicago). Instructor 1.17.0 returned a validated Pydantic object from a local 3B model in 3 of 3 trials (structured extraction, not a tool call); cold uv install 1.4 s. Test environment: Linux x86_64 virtual machine (Debian 13), 8 vCPU, 15 GiB RAM, no GPU; no vendor account, login or API key. Models were local (Qwen2.5-3B-Instruct (Q4_K_M) through Ollama 0.40.11); we tested the tool, not any vendor's hosted models. Not tested: Any vendor-hosted model: everything ran on a local 3B model, so tool-call reliability with frontier models is not measured here; Multi-agent flows, memory, long conversations, streaming, tracing and cloud deployment; Anything beyond one tool and one question; three trials is a smoke test, not a benchmark. These are smoke tests, not a benchmark. Full log: /tests/instructor.

Evidence
Tested (smoke test, 2026-10-25, local 3B model (Qwen2.5-3B)). A smoke test is not a benchmark.
What we ran
Instructor 1.17.0 returned a validated Pydantic object from a local 3B model in 3 of 3 trials (structured extraction, not a tool call); cold uv install 1.4 s.
Model used
Qwen2.5-3B-Instruct (Q4_K_M) through Ollama 0.40.11
Environment
Linux x86_64 virtual machine (Debian 13), 8 vCPU, 15 GiB RAM, no GPU; no vendor account, login or API key
Not tested
Any vendor-hosted model: everything ran on a local 3B model, so tool-call reliability with frontier models is not measured here; Multi-agent flows, memory, long conversations, streaming, tracing and cloud deployment; Anything beyond one tool and one question; three trials is a smoke test, not a benchmark.
Date
(America/Chicago)

Best for

Key facts

Instructor is a Python library that turns model responses into validated Pydantic objects. It is MIT-licensed; v1.17.0 was tagged on 2026-09-09.

  • License: MIT (per GitHub)
  • Platforms: Python
  • Pricing model: Open source
  • Free tier: Open source, free to use
  • Paid plans: Hosted or enterprise offerings not read
  • GitHub: 567-labs/instructor: 14,063 stars on 2026-10-25, last push 2026-10-20
  • Latest release we saw: v1.17.0, 2026-09-09
  • How to start: uv pip install instructor and follow the docs.
  • v1.17.0 (2026-09-09): the newest release on the repository when we read it on 2026-10-25.

How to set it up: Get started with Instructor

Install steps are exactly what we ran on 2026-10-25; the test log lists what was and was not covered.

  1. Install

    A light install in our measurement.

    uv pip install instructor

  2. Wrap your client

    Patch an OpenAI-compatible client and pass a Pydantic response_model.

    client = instructor.from_openai(OpenAI(base_url="http://127.0.0.1:11434/v1", api_key="x"))

Pros and cons

Pros

  • MIT license
  • Small install
  • Worked against a local OpenAI-compatible server

Cons

  • Structured output only: not an agent framework
  • We tested one extraction, not retries or streaming

Alternatives and comparisons

  • Pydantic AI: Python agent framework built on Pydantic types, covering agents, realtime voice, image generation and embeddings.
  • DSPy: Python framework for programming (rather than prompting) language models, with optimisers for prompts and weights.
  • LiteLLM: Python SDK and proxy that gives one OpenAI-style interface to many model providers.
  • Langflow: MIT-licensed visual builder for AI agents and MCP servers with a very large GitHub following.

FAQ

What is Instructor?

A Python library that makes language models return validated Pydantic objects.

Did Instructor work with a local model in our test?

Yes: 3 of 3 trials returned a valid structured result from a local 3B model. See /tests/instructor.

How big is the Instructor install?

About 49 MB and 41 packages with uv on our machine in 1.4 s from a cold cache.

What we have not verified

  • Retries, validation failures and streaming: not run
  • Behaviour with hosted models

Sources

Last verified · Agentic Weekly

  • Langflow: MIT-licensed visual builder for AI agents and MCP servers with a very large GitHub following.
  • Botpress: Customer-support AI agent platform with a free plan and conversation-based paid plans.
  • Gumloop: Agent builder for teams; plans start at $37 per month with 20k included credits.

How to cite this page

Agentic Weekly, "Instructor v1.17.0: what's new, verdict & setup", last updated 2026-10-25, https://theagenticweekly.com/tools/instructor

Keep the Reported or Tested label when you quote a verdict. Plain-text version: https://theagenticweekly.com/tools/instructor.md. More for AI assistants: /ai, /tools.json, /llms.txt.