Skip to content
Agents tracked: 501 Releases (7d): 777 Agent status: no reported issuesSince last limit reset: Claude 10d · Codex 8d For AI agents: Markdown Updated Nov 21, 2026

STORM by Stanford OVAL

An open-source system from Stanford that researches a topic through web search and multi-perspective question asking, then writes a full-length, Wikipedia-style report with citations.

AI research agents Open-source research and report-writing agentOpen source · MITFree to start

Use it in: API / SDK Web app Self-hosted Works with: Own API key Multi-agent

Price
Free (OSS)
free and open source
Popularity
#9 in Research
Gid Score 22 / 100 · #216 of 255 ranked AI agents
Trend
Steady
downloads down 3.5% over 30 days · up 87.9% last week
Development
Quiet lately
latest 1.1.0, 22mo ago
Community
31.9K GitHub stars
404 package downloads a week

Usage data: latest readings Nov 19 to Nov 21, 2026, last checked Nov 21, 06:26 UTC. Date of each number

What STORM does

STORM is an open-source research and report-writing agent from Stanford OVAL, first released in 2024. You use it through an SDK, a web app and self-hosted. Its documentation covers subagents or parallel agents and your own API key. It can run models from OpenAI, Azure and any OpenAI-compatible endpoint. It is used from Python. It is free and open source (MIT); you pay only for the models it calls.

Key facts

MakerStanford OVAL
TypeOpen-source research and report-writing agent
First released2024
LicenseMIT
InterfacesSDK, Web app, Self-hosted
Main languagePython
Open issues and PRs120

Pricing

Free open-source library plus a free research demo site; self-hosting users pay their own model and search costs.

Source: official pricing page, checked Oct 6, 2026. Prices change often; confirm before you buy.

Compare prices of all agents

Quick answers

  • Free to start?Yes
  • Open source?Yes (MIT)
  • Where you use itSDK, Web app, Self-hosted
  • Model providersopenai, azure, any OpenAI-compatible
  • Subagents / multi-agent?Yes
  • Feature comparisonAll agents
  • Closest alternativeBettaFish · all alternatives

What's new in STORM: January 2025

Full monthly summary

January brought expanded model support and new search capabilities to STORM. The release adds flexibility for users working with different language model providers.

  • STORM now works with any language model supported by LiteLLM, greatly expanding model options.
  • Azure AI Search is now available as a retrieval option.
  • Fixed a bug in the Serper retriever.

1 release this month · summarised from the official release notes · all months

Get started with STORM

PyPIpip install knowledge-storm

These commands use the packages tracked on this page. The vendor may recommend a different installer; see the official documentation.

Compare STORM with another agent

vs
# Agent Gid Score Downloads 7d 7d 30d Stars Latest release Price
71
BettaFish666ghj (community project)
432/4 — — — 42.8K+11/day 3.0.011mo ago Free (OSS)
72
Paper Orchestra NewGoogle Research
432/4 — — — 5,493+757/day — Free (OSS)
84
Autoresearchkarpathy
41 — — — 100K+52/day — Free (OSS)
88
HorizonSheng Qiang
402/4 — — — 14.9K+112/day — Free (OSS)
93
DexterVirat Singh
402/4 — — — 28K+22/day 1.1.059d ago Free (OSS)
155
GPT ResearcherAssaf Elovic and contributors
32 9,793 up 6.3% down 13.2% 30.8K+20/day 3.8.012d ago Free (OSS)

Tools used with STORM

All tools

What 10 public GitHub projects that depend on STORM use with it, from their dependency files.

  1. pytest60.0%
  2. NumPy50.0%
  3. python-dotenv50.0%
  4. python-multipart40.0%
  5. FastAPI40.0%
  6. LangChain40.0%
  7. Uvicorn40.0%
  8. Anthropic Python SDK40.0%
  9. Pydantic40.0%
  10. Streamlit40.0%

Used far more often here than in AI-built projects overall: what sets STORM projects apart.

  1. Extra Streamlit Components20.0% · 1251.0×
  2. Voyage AI (Python)20.0% · 417.0×
  3. AutoGen20.0% · 417.0×
  4. Gymnasium20.0% · 250.2×
  5. Unidecode20.0% · 208.5×
  6. Mutagen10.0% · 208.5×
  7. Trafilatura20.0% · 156.4×
  8. toml (Python)20.0% · 125.1×

Ratings from people who use STORM

No ratings yet. Used STORM? Rate it and say what it is good and bad at.

What changed: recent STORM releases

Monthly summary · Full changelog

1.1.0
  • #309 STORM & Co-STORM now accept language models that are supported by LiteLLM
  • #236 Serper retriever bug fix
  • #198 Add Azure AI search
1.0.0
  • [2024/09] Co-STORM is now live and fully integrated into the knowledge-storm Python package. Try it out by upgrading the package today!
  • [2024/09] Our paper on Co-STORM has been accepted to EMNLP 2024! You can read it here.
  • Co-STORM is now integrated into the knowledge-storm package. Check out the API documentation for more details.
  • We introduce the Agent interface in Co-STORM, providing a unified framework for defining different LM agent policies for information seeking and knowledge curation.
  • Co-STORM LLM Experts: These agents ground their responses in external knowledge sources and ask follow-up questions based on the discourse.
  • Moderator Agent: Guides the conversation by generating insightful questions, drawing attention to discovered but underexplored areas.
Earlier releases
0.2.4
  • GoogleModel
  • DeepSeekModel
  • Improve VectorRM by decoupling the vector store construction from the retriever class.
0.2.3
  • You can now install storm via pip install knowledge-storm.
  • We update the README.md to explain our current API.
  • If you prefer using the source code, note that pulling the new release will lead to breaking change as we rename src/ to knowledge_storm/.
0.2.0
  • Add VectorRM to support grounding on user-provided documents, complementing existing support of search engines (YouRM, BingSearch).
  • We provide an example script here and detailed instructions here.
  • We release a new UI for users to directly interact with STORM, available at This update brings improved stability and a new feature of discovering popular topics.
  • We release demo light for developers, a minimal user interface built with streamlit framework in Python, handy for local development and demo hosting
  • In-person talk in San Francisco: Our team member will give an invited talk at Wikipedians Meetup @ San Francisco on July 11. Come to chat with us in person!
  • Upcoming major update: A major update supporting human-AI collaboration mode will be released over the summer. Stay tuned!
0.1.0
  • GPT family
  • Claude family
  • VLLM client
  • TGI client
  • Together Client

STORM prices over time

Token prices of OpenAI modelsnewest OpenAI model in each family, output price per 1M tokens
OpenAI FlagshipOpenAI Mini

STORM popularity and usage data

STORM was downloaded 404 times from PyPI in the week ending Nov 19, 2026, up 87.9% on the week before and down 3.5% over the last 30 days compared with the 30 before. The GitHub repository has 31.9K stars, 289 of them added in the last 30 days. The latest stable release is 1.1.0, published Jan 23, 2025.

Downloads per dayPyPI, 7-day average 58/day
GitHub starsrunning total 31.9K

Download the raw daily series: storm.csv

Gid Score breakdown

Adoption0
Community63
Attention—
Momentum46

Gid Score shows public popularity and activity (usage, GitHub stars, mentions, growth), not the quality of the product. Each component is scored 0 to 100 from public signals; a dash means no data for it. How the Gid Score works.

How fresh these numbers are

NumberLatest readingLast checked
Package downloadsNov 19, 2026Nov 21, 06:24 UTC
GitHub starsNov 21, 2026Nov 21, 06:26 UTC

We fetch every source once a day. Registries publish downloads and page views a day or two late, so their latest reading trails the check. A reading more than 4 days old is marked out of date; the Gid Score still uses it until a newer one arrives.

STORM FAQ

How much does STORM cost?

Free open-source library plus a free research demo site; self-hosting users pay their own model and search costs.

Is STORM open source?

Yes. STORM is open source, released under the MIT license.

How popular is STORM?

STORM was downloaded 404 times from PyPI in the week ending Nov 19, 2026, up 87.9% on the week before and down 3.5% over the last 30 days compared with the 30 before. The GitHub repository has 31.9K stars, 289 of them added in the last 30 days. The latest stable release is 1.1.0, published Jan 23, 2025.

What is the latest version of STORM?

The latest stable release we track is 1.1.0, published on Jan 23, 2025.

What are the alternatives to STORM?

The closest alternatives in the same category by Gid Score are BettaFish, Paper Orchestra, Autoresearch, Horizon.

For the makers of STORM badges, corrections

Add a live badge to your README or site:

AgentGid: #9 in Research Gid Score: 22/100 downloads: 404/week ▼4% 30d

[![AgentGid: #9 in Research](https://agentgid.com/badges/storm-rank.svg)](https://agentgid.com/agents/storm/)
[![Gid Score: 22/100](https://agentgid.com/badges/storm-score.svg)](https://agentgid.com/agents/storm/)
[![downloads: 404/week ▼4% 30d](https://agentgid.com/badges/storm-downloads.svg)](https://agentgid.com/agents/storm/)

Something wrong or outdated on this page? Send a correction with a link to the official source. More for makers.

Where these numbers come from

PyPI: knowledge-storm. GitHub: stanford-oval/storm. See data sources for how each one is collected.

Product facts and pricing were checked against: github.com, storm.genie.stanford.edu.