aiograpi - Asynchronous Instagram API for Python

May 13, 2026 · View on GitHub

⚠️ Telegram support group moved to https://t.me/aiograpi_support — the previous @instagrapi group has been restricted by Meta and is no longer maintained.

If you want to work with aiograpi (business interests), we strongly advise you to prefer HikerAPI project. However, you won't need to spend weeks or even months setting it up. The best service available today is HikerAPI, which handles 4–5 million daily requests, provides support around-the-clock, and offers partners a special rate. In many instances, our clients tried to save money and preferred aiograpi, but in our experience, they ultimately returned to HikerAPI after spending much more time and money. It will be difficult to find good accounts, good proxies, or resolve challenges, and IG will ban your accounts.

The aiograpi more suits for testing or research than a working business!

Video uploads can use a built-in MP4 metadata parser when you provide thumbnail=.... Automatic thumbnail generation, StoryBuilder, and video/audio composition still need executable ffmpeg; Android/Pydroid users should see Pydroid and ffmpeg.

We recommend using our services:

PyPI Python License Package Docs ZeroVer

Features:

  • Getting public data of user, posts, stories, highlights, followers and following users
  • Getting public email and phone number, if the user specified them in his business profile
  • Getting public data of post, story, album, Reels, IGTV data and the ability to download content
  • Getting public data of hashtag and location data, as well as a list of posts for them
  • Getting public data of all comments on a post and a list of users who liked it
  • Management of proxy servers, mobile devices and challenge resolver
  • Login by username and password, sessionid and support 2FA
  • Managing messages, reactions and threads for Direct and attach files
  • Download and upload a Photo, Video, IGTV, Reels, Albums, Stories and Trial Reels
  • Work with Users, Posts, Comments, Insights, Collections, Location and Hashtag
  • Insights by account, posts and stories
  • Like, following, commenting, editing account (Bio) and much more else

Asynchronous Instagram Private API wrapper without selenium. Use the most recent version of the API from Instagram, which was obtained using reverse-engineering with Charles Proxy and Proxyman.

Support Python >= 3.10

For any other languages (e.g. C++, C#, F#, D, Golang, Erlang, Elixir, Nim, Haskell, Lisp, Closure, Julia, R, Java, Kotlin, Scala, OCaml, JavaScript, Crystal, Ruby, Rust, Swift, Objective-C, Visual Basic, .NET, Pascal, Perl, Lua, PHP and others), I suggest using instagrapi-rest

Support chat in Telegram: https://t.me/aiograpi_support and GitHub Discussions

Features

  1. Performs Web API or Mobile API requests depending on the situation (to avoid Instagram limits)
  2. Login by username and password, including 2FA and by sessionid (and uses Authorization header instead Cookies)
  3. Challenge Resolver have Email and SMS handlers
  4. Support upload a Photo, Video, IGTV, Reels, Albums and Stories
  5. Support work with User, Media, Comment, Insights, Collections, Location (Place), Hashtag and Direct Message objects
  6. Like, Follow, Edit account (Bio) and much more else
  7. Insights by account, posts and stories
  8. Build stories with custom background, font animation, link sticker and mention users
  9. Account registration and captcha passing will appear

Versioning policy — we stay in 0.x

aiograpi follows ZeroVer. We will not ship 1.0, because it would be a misleading promise. Instagram's private API rotates doc_ids, deprecates endpoints, and changes response shapes without notice — we cannot guarantee a stable API surface across versions. Pretending otherwise hurts users who'd take "1.0" to mean "safe to pin and forget".

What you can rely on instead:

  • Breaking changes are flagged in the CHANGELOG with a Breaking section per affected release. No silent breakage.
  • Deprecated methods stay around for ≥2 minor releases with DeprecationWarning before removal — you'll get loud warnings, not surprise AttributeErrors.
  • Live CI smoke runs on every push: tests/live/smoke.py against a real account through a real proxy. If we ship something that breaks the basic happy path, CI catches it.
  • Migration Guide at docs/migration.md — every breaking change documented with before/after.

What's new in 0.6.x through 0.9.x

  • Sync with instagrapi 2.6.1 — current Android app profile defaults, override_app_version constructor support, Trial Reels, current Reel rupload flow, Reel pin/unpin, feed photo/carousel music, music Notes, archive readers, tagged media pagination, Direct reactions, thread title updates, message request helpers, single-message lookup, and Direct unsend.
  • Android/Pydroid-friendly video uploads — when you pass thumbnail=..., aiograpi can read MP4 dimensions/duration without importing MoviePy/ffmpeg. If thumbnail generation is needed, the error now explains how to install ffmpeg or set IMAGEIO_FFMPEG_EXE.
  • Modern dev tooling — uv.lock, Ruff formatting/checks, updated test pins, and an upstream sync tracking workflow.
  • Sync with instagrapi 2.4.4 — every mixin and infrastructure module ported, plus three new mixins: ExploreMixin, FundraiserMixin, and opt-in CaptchaHandlerMixin.
  • doc_id GraphQL primitive — Client.public_doc_id_graphql_request(doc_id, variables) and Client.private_graphql_query_request(...) for the new i.instagram.com/graphql/query surface IG migrated to. New high-level methods: user_info_v2_gql, user_info_by_username_v2_gql, private_graphql_followers_list / following_list / clips_profile / inbox_tray_for_user, private_graphql_memories_pog / realtime_region_hint / top_audio_trends_eligible_categories, plus fbsearch_keyword_typeahead / fbsearch_typeahead_stream / fbsearch_item, feed_user_stream_item, media_comment_infos. All live-verified.
  • Pure helpers go sync (breaking from 0.0.x): media_pk_from_code, media_code_from_pk, media_pk, share_info, share_code_from_url, share_info_by_url, highlight_pk_from_url, handle_challenge_result, challenge_resolve_new_password_form no longer require await. See Migration Guide for the full list.
  • CI publish-on-tag with PyPI trusted publishing — push a version tag, GitHub Actions builds, publishes, and creates a release.

Full per-release notes: CHANGELOG.md. Migrating from 0.0.x? See the Migration Guide.

Installation

pip install aiograpi

Basic Usage

from aiograpi import Client

cl = Client()
await cl.login(ACCOUNT_USERNAME, ACCOUNT_PASSWORD)

user_id = await cl.user_id_from_username(ACCOUNT_USERNAME)
medias = await cl.user_medias(user_id, 20)

Session Persistence

Logging in fresh on every run is the fastest way to get your account flagged. Persist the session and reuse it:

from aiograpi import Client

cl = Client()
await cl.login(USERNAME, PASSWORD)
cl.dump_settings("session.json")

# reload later without entering credentials again
cl = Client()
cl.load_settings("session.json")
await cl.login(USERNAME, PASSWORD)

If you want explicit control over the loaded session object:

from aiograpi import Client

cl = Client()
cl.set_settings(cl.load_settings("session.json"))
await cl.login(USERNAME, PASSWORD)

Login by sessionid

from aiograpi import Client

cl = Client()
await cl.login_by_sessionid("<your_sessionid>")

login_by_sessionid() is a lightweight compatibility path. For long-lived automation prefer the regular login() → dump_settings() → load_settings() / set_settings() flow.

Typical Tasks

List and download another user's posts

target_id = await cl.user_id_from_username("target_user")
posts = await cl.user_medias(target_id, amount=10)
for media in posts:
    await cl.photo_download(media.pk)

Search a location by name

places = await cl.location_search("Times Square")
print(places[0].name, places[0].pk)

Followers via the new private GraphQL surface

import uuid

data = await cl.private_graphql_followers_list(
    user_id="25025320",
    rank_token=str(uuid.uuid4()),
)
# Raw GraphQL envelope: {"data": {...}, "status": "ok", ...}

See Private GraphQL & doc_id for the full new-mobile-API surface (followers, clips, search, inbox).

Additional example (story upload with mentions / hashtags / media) 
from aiograpi import Client
from aiograpi.types import StoryMention, StoryMedia, StoryLink, StoryHashtag

cl = Client()
await cl.login(USERNAME, PASSWORD, verification_code="<2FA CODE HERE>")

media_pk = await cl.media_pk_from_url('https://www.instagram.com/p/CGgDsi7JQdS/')
media_path = await cl.video_download(media_pk)
subzeroid = await cl.user_info_by_username('subzeroid')
hashtag = await cl.hashtag_info('dhbastards')

await cl.video_upload_to_story(
    media_path,
    "Credits @subzeroid",
    mentions=[StoryMention(user=subzeroid, x=0.49892962, y=0.703125, width=0.8333333333333334, height=0.125)],
    links=[StoryLink(webUri='https://github.com/subzeroid/aiograpi')],
    hashtags=[StoryHashtag(hashtag=hashtag, x=0.23, y=0.32, width=0.5, height=0.22)],
    medias=[StoryMedia(media_pk=media_pk, x=0.5, y=0.5, width=0.6, height=0.8)]
)

Documentation

Contributing

List of contributors

Releasing

Releases are automated via the publish.yml GitHub Actions workflow with PyPI trusted publishing. To cut a new release:

  1. Bump version = in pyproject.toml.
  2. Add a section to CHANGELOG.md.
  3. Commit and push to main.
  4. Tag: git tag -a 0.x.y -m "Release 0.x.y" && git push origin 0.x.y.

The workflow then builds sdist + wheel, verifies the tag matches pyproject.toml, publishes to PyPI, and creates the GitHub release with both artefacts attached — no API tokens needed.