Usage may lead to the loss of the Instagram account being used. Please see Limitations and Ethical Considerations before using this tool.
Description
Instaloader lets you save Instagram content for documentation and analysis. It can download posts, videos, Stories, Highlights, Reels, and profile pictures. It also exports captions, timestamps, and geotags so you keep context with the media.
Investigators typically use it to archive a profile, collect evidence before takedowns, or fetch posts for later review. Logged-in mode provides access to your feed, saved posts, and followed accounts, where permitted.
You can run it from the command line or script it in Python. It supports resuming interrupted downloads and only grabs new items to keep an archive up to date.
Media Downloading:
Downloads photos and videos from public and private Instagram profiles, hashtags, Stories, feeds, saved media, long-form videos (formerly IGTV), and Reels (--reels flag).
It captures metadata, including captions, comments, geotags (such as Google Maps links), timestamps, and more.
Allows downloading from private profiles if you follow the account or provide valid credentials.
Offers optional parameters to limit downloads (e.g., specific sidecar slides, date ranges, post filters).
Automation and Customization:
Automatically detects profile name changes and renames the target directory accordingly.
It supports resuming downloads from previously interrupted sessions, both at the command-line level (JSON resume files) and via the Python API.
Offers fine-grained control over what to download through various filters (e.g., specific date ranges, post types) and customization of directory and filename patterns.
Can be configured for automated tasks (e.g., cron jobs) using --quiet mode and persistent session files.
Recommended one‑liner: reliably works on Linux, macOS, WSL, and Windows (PowerShell)
python3 -m pip install --upgrade instaloader
For isolated environments:Pipx is a tool to install and run end-user Python applications in isolated virtual environments. It's an automation layer over pip and venv specifically for Python CLI applications. That means each app gets its own virtual environment (venv) to avoid conflicts with other installed scripts while still enabling console scripts globally.
pipx install instaloader
For users ofAnaconda via conda-forge:Anaconda is an open-source distribution of the Python and R languages for data science that simplifies package management and deployment by using the conda environment manager to resolve dependencies and avoid conflicts. Conda is the cross-platform package and environment manager included in Anaconda; it supports multiple channels for sourcing packages. “Conda-forge” is a community-driven packaging effort and GitHub organization that maintains recipes for building conda packages. Built distributions are published to anaconda.org/conda-forge and can be installed into any conda-based environment, providing an open-source alternative to the default Anaconda channel.
Just use your terminal or PowerShell, and make sure that the environment you want to install (just substitute "<env-name>" with "myenvironment", the name is up to you), this package is activated:
git clone https://github.com/instaloader/instaloader.git
cd instaloader
python setup.py install
When to use this option
Need the latest dev version: grab unreleased fixes or features that aren’t on PyPI yet.
Plan to tweak the code: cloning the repo puts the full source on your machine so you can patch or extend Instaloader, then reinstall instantly.
Audit / offline install: You can inspect every line before it touches your system, or vendor a specific commit for reproducibility.
What the three commands do
git clone … downloads the Instaloader source code into a new folder.
cd instaloader moves into that folder.
python -m pip install . (or python setup.py install) builds and installs the package from the source you just checked out.
&#xNAN;Tip: run this inside a virtual-env so it won’t clash with other Python tools.
Example Use Cases: CLI
Download & continuously update one public profile
# First run (creates folder instagram/)
instaloader instagram
# Daily cron job – only grab *new* posts & stories
instaloader --login YOUR_USER --stories --fast-update instagram
Archive only the last 100 posts with the hashtag "#ukraine"
instaloader --count 100 "#ukraine"
Fetch Stories only (no posts) from multiple private accounts you follow
Download all available posts between two dates (server‑side filter)
# Posts from 1 Jan 2024 – 31 Dec 2024
instaloader --login YOUR_USER \
--post-filter="date_utc >= datetime(2024,1,1) and date_utc < datetime(2025,1,1)" \
instagram
The filter expression syntax is Python‑like and evaluated per post.
Resume an interrupted job
instaloader --resume-prefix myrun instagram
Instaloader writes a JSON checkpoint so the next invocation restarts exactly where it left off. You can define the --resume-prefix and --no-resume flags for resuming interrupted download loops. When resuming is enabled, Instaloader writes a compressed JSON “checkpoint” file into the target directory, containing all information needed to pick up where it left off. (https://instaloader.github.io/cli-options.html)
Example Use Cases: Python
All snippets require pip install instaloader and should live in their own virtual environment or notebook.
Download the N most recent posts from a profile
from itertools import islice
import instaloader
PROFILE = "instagram"
MAX_POSTS = 10
L = instaloader.Instaloader()
L.context.log("Logging in…")
L.load_session_from_file("my_user") # reuse saved session, or:
# L.login("my_user", "PASSWORD")
posts = instaloader.Profile.from_username(L.context, PROFILE).get_posts()
for post in islice(posts, MAX_POSTS):
L.download_post(post, target=PROFILE)
print("Done.")
List (and optionally download) stories of accounts you follow
import instaloader, datetime as dt
L = instaloader.Instaloader()
L.load_session_from_file("my_user")
for story in L.get_stories(): # generator of Story objects
print(f"{story.owner_username} has {story.itemcount} items")
for item in story.get_items():
age = dt.datetime.utcnow() - item.date_utc
if age.days < 1: # only today’s stories
L.download_storyitem(item, f"stories_{story.owner_username}")
"""
Download posts from multiple profiles (Jan–Mar 2025) and write core
metadata (caption, likes, timestamp) to posts.csv
"""
import csv, datetime as dt, instaloader
from dateutil.tz import UTC
PROFILES = ["cnn", "bbcnews", "dwnews"]
START, END = dt.datetime(2025,1,1,tzinfo=UTC), dt.datetime(2025,4,1,tzinfo=UTC)
L = instaloader.Instaloader(dirname_pattern="downloads/{target}", quiet=True)
L.load_session_from_file("YOUR_USER")
rows = []
for username in PROFILES:
profile = instaloader.Profile.from_username(L.context, username)
for post in profile.get_posts():
if post.date_utc < START:
break # posts are reverse-chronological → done
if START <= post.date_utc < END:
L.download_post(post, target=username)
rows.append({
"user": username,
"shortcode": post.shortcode,
"date": post.date_utc.isoformat(),
"likes": post.likes,
"caption": post.caption or ""
})
with open("posts.csv", "w", newline="", encoding="utf-8") as f:
writer = csv.DictWriter(f, fieldnames=rows[0].keys())
writer.writeheader(); writer.writerows(rows)
print(f"Exported {len(rows)} rows to posts.csv")
Detect new followers & unfollowers overnight
"""
Run daily via cron; prints who followed / unfollowed you since yesterday.
"""
import json, instaloader, pathlib, datetime as dt
TS = dt.date.today().isoformat()
STATE = pathlib.Path("followers_history.json")
L = instaloader.Instaloader(quiet=True); L.load_session_from_file("YOUR_USER")
me = instaloader.Profile.from_username(L.context, "YOUR_USER")
today = sorted({f.username for f in me.get_followers()})
if STATE.exists():
yesterday = json.loads(STATE.read_text())
else:
yesterday = []
new = sorted(set(today) - set(yesterday))
unfollow = sorted(set(yesterday) - set(today))
print("New followers:", new if new else "—")
print("Unfollowers:", unfollow if unfollow else "—")
STATE.write_text(json.dumps(today, indent=2))
Generate a hashtag frequency list from one username
"""
Build hashtag counts across the latest 400 posts of a target profile.
Output TSV sorted by frequency.
"""
import re, collections, instaloader, itertools, csv
TARGET = "nasa"
MAX_POSTS = 400
hashtag_re = re.compile(r"#(\w+)")
freq = collections.Counter()
L = instaloader.Instaloader(quiet=True)
for post in itertools.islice(instaloader.Profile.from_username(L.context, TARGET).get_posts(), MAX_POSTS):
freq.update(hashtag_re.findall(post.caption or ""))
with open(f"{TARGET}_hashtags.tsv", "w", newline="", encoding="utf-8") as f:
writer = csv.writer(f, delimiter="\t")
writer.writerow(["hashtag", "count"])
for tag, count in freq.most_common():
writer.writerow([tag.lower(), count])
print("Top 10:", freq.most_common(10))
Rate-limit-aware bulk hashtag scraper with progress-bar (via tqdm)
"""
Scrape 1 000 #ukraine posts, obeying a minimum 2-second delay between requests.
"""
import time, itertools, instaloader, tqdm
HASHTAG = "ukraine"
TARGET = f"hashtag_{HASHTAG}"
MAX_ITEMS = 1_000
MIN_DELAY = 2.0 # seconds
L = instaloader.Instaloader(quiet=True)
posts = itertools.islice(instaloader.Hashtag.from_name(L.context, HASHTAG).get_posts(), MAX_ITEMS)
for post in tqdm.tqdm(posts, total=MAX_ITEMS):
start = time.time()
try:
L.download_post(post, target=TARGET)
except instaloader.exceptions.QueryReturnedNotFoundException:
continue # skip removed posts
# enforce polite delay
elapsed = time.time() - start
if elapsed < MIN_DELAY:
time.sleep(MIN_DELAY - elapsed)
How to integrate these scripts
One-shot analysis: run directly from your project folder after activating the virtual-env.
Scheduled jobs: wrap in a shell script and call via cron (crontab -e) or Windows Task Scheduler.
Notebook workflow: import functions from each script to mix Scraping + Pandas analysis in Jupyter.
Feel free to remix flags (e.g., add download_comments=True) or combine ideas (geofilter + hashtag counter) to suit your investigation.
CLI-to-Python “cheat sheet” (v 4.14.1, 12 Jul 2025)
Below you’ll find every current command-line flag grouped by purpose, with the canonical Python call or argument that achieves the same effect.
Where the Instaloader constructor accepts a parameter mapped 1-to-1 to the flag, it is shown like Instaloader(download_pictures=False).
When the flag controls behaviour per download call, the mapping is shown next to the relevant high-level method (e.g. download_profiles(..., fast_update=True)).
Tip: create one configured loader and reuse it:
from instaloader import Instaloader
L = Instaloader(dirname_pattern="{target}", quiet=True, download_geotags=True)
Open full length "cheat-sheet"...
CLI flag
What it does
Python equivalent
--login USER, --password PASS
Interactive/password login & save session
L.login(user, passwd) or L.interactive_login(user)
(= smallest amount of Python code you need to get a working command-line interface whose commands exactly mirror (i.e., have feature parity with) your underlying API or library.)
"""
Replicates: instaloader --login USER --quiet --stories --no-pictures --geotags profile1 profile2
"""
import instaloader, itertools
L = instaloader.Instaloader(
quiet=True,
download_pictures=False,
download_geotags=True)
L.login("USER", "PASSWORD") # or L.load_session_from_file("USER")
targets = ["profile1", "profile2"]
profiles = {instaloader.Profile.from_username(L.context, t) for t in targets}
L.download_profiles(profiles,
stories=True,
posts=True) # posts=True because we only disabled pictures
Special Relationship: Instaloader & Bellingcat Auto Archiver
When you're working with the Bellingcat Auto Archiver, you may be asked to provide instaloader.session
instaloader.session is a file created by the Instaloader tool (which Auto Archiver uses under the hood for Instagram content). It is a small data file that stores your login session (including cookies) for Instagram so that you don’t have to log in every time you run Instaloader or the Auto Archiver.
Why It Exists
Instagram requires users to log in for certain types of data access (especially if you need to see private or protected content). When you enter your username and password in Instaloader for the first time, the tool authenticates with Instagram and fetches a session token (a cookie).
Instaloader then saves that session token locally to a file called instaloader.session.
How It Is Created
The first time you run Instaloader (or Auto Archiver with the instagram_archiver step enabled) and attempt to archive from a private Instagram account or any content that requires login; Instaloader will prompt you for Instagram credentials.
After successful login, a file named instaloader.session appears in the working directory (or wherever the tool is configured to store session data).
What’s Inside
Internally, it contains cookies that represent your authenticated session on Instagram. These cookies let Instaloader reuse your login, so you don’t have to type your username and password on each run.
It also ensures Instagram recognizes you in the same way as if you logged in via your browser.
Where to Store It
For Auto Archiver, best practice is to place instaloader.session (once generated) in your secrets/ folder. That way, the Docker container or local script knows where to find your session, and you’re not repeatedly asked for credentials.
Remember to keep this file private and avoid committing it to version control (e.g., GitHub). Anyone with access to it could theoretically use your Instagram session.
Expiration & Re-authentication
Sometimes, Instagram invalidates session cookies (e.g., for security or if you change your password). If that happens, Instaloader may prompt you to log in again and create a newinstaloader.session file.
If you notice Instagram's archiving stops working, you might need to delete the old session file and let Instaloader recreate it with fresh credentials.
Security Considerations
Treat instaloader.session as if it’s your actual Instagram login. Anyone who obtains a valid session file might have the same access to your account.
Use a dedicated Instagram account for archiving whenever possible (instead of a personal one), and store that account’s session file in a secure place.
Cost
Instaloader is a free, open-source project.
Level of difficulty
While basic usage (e.g., downloading all posts from a profile) is straightforward, some command-line or scripting familiarity helps when applying advanced filters, scheduling automation, or handling custom workflows.
Requirements
System Requirements
Instaloader is platform-independent and can run on most operating systems (Linux, macOS, Windows, etc.).
Python Environment
Python 3.9 or higher (tested up to 3.13, 3.8 support was dropped in v4.14)
Primary dependencies (installed automatically with pip):
requests Requests lets you send HTTP/1.1 requests effortlessly, with automatic query-string encoding, JSON support, and more
lxml is a feature-rich and easy-to-use library for processing XML and HTML in the Python language, with good performance, and memory efficiency,
BeautifulSoup4 Python package for parsing HTML/XML (including malformed markup) into a navigable parse tree for data extraction. Provides idiomatic tools for iterating, searching, and modifying the parse tree (optional; only needed for certain features)
Access Requirements
Anonymous Access: Instaloader can scrape public profiles, hashtags, and posts without login.
Logged-In Access: To scrape private profiles or retrieve user-specific data (e.g., user feed, saved posts), you must log in with valid Instagram credentials. Instaloader stores a session file to avoid repeated logins.
Limitations
Heavy scripted use can trigger temporary locks or permanent bans.
Technical Barriers:
Command-line usage can pose a challenge for non-technical users.
Using the Python module requires basic programming knowledge.
Rate Limits and Restrictions:
Instagram actively attempts to detect and restrict automated scraping. Even moderate usage of Instaloader can lead to security warnings or temporary locks on your account.
Instagram imposes strict rate limits that can trigger 429 Too Many Requests errors if requests exceed certain thresholds.
Using proxies or VPNs may result in stricter rate limits for anonymous scraping.
Instagram does not publicly disclose the exact thresholds for the web endpoints, so the limits can change without notice. Based on user experiences, the thresholds have become much stricter in recent years. For example, one user noted that previously, you could make on the order of ~200 requests per minute without issues. Current anecdotalceiling is closer to 1–2 requests / 30s for unauthenticated scrapes; thresholds vary by endpoint and change frequently. Community reports now place anonymous limits around 1–2 requests every 30 seconds, sometimes lower. In other words, Instagram dramatically lowered the allowed request rate for scraping. Another Instaloader user observed getting a 429 after analyzing just 2–3 posts in a row, even from different machines, indicating a very low threshold in effect. The exact limit may vary over time or by content type – for instance, story downloads seem to trigger limits faster than regular posts (Instagram appears to “heavily rate-limit stories” requests).
Checkpoint/401 (login verification prompts) errors are common; importing browser cookies (using -b chrome or similar) is a recommended workaround.
Data Integrity:
Instaloader lacks built-in hashing to verify the authenticity of downloaded content.
Ethical and Legal Constraints:
Scraping private profiles or sensitive data without authorization can violate ethical norms and Instagram’s Terms of Service.
Disclaimer: Instaloader is unaffiliated with Instagram and comes with no warranty. Always review local regulations and platform rules before usage.
Rate Limiting Mitigation Strategies
If you encounter 429 errors or want to avoid them, consider these practices:
Stay Logged In: Always use --login with Instaloader so that you leverage a logged-in session (and reuse a session file). This not only grants access to more content but may raise the rate limit ceiling compared to anonymous use. The Instaloader team advises keeping the session file to avoid re-login each run which prevents frequent login attempts that could trigger Instagram’s security checks.
Do Not Parallelize or Re-run Quickly:Run a single instance of Instaloader at a time and avoid launching multiple scrapes in parallel. Also, avoid scheduling runs back-to-back without sufficient delay. Give some breathing room between Instaloader runs (or between different scraping scripts) so that Instagram’s counters can reset. For example, if you update a profile archive, you might schedule it hourly or daily, not every minute.
Limit Request Frequency: Instaloader already inserts delays between requests, but if you still hit 429s, you may need to slow down further. You can insert additional sleep/delay in your scraping logic or use a custom RateController. One user found success by randomizing delays between actions: e.g., waiting 10–30 seconds between different profiles or hashtag scrapes, 1–3 seconds between individual post downloads, and 3–10 seconds between story fetches. These pauses mimic human browsing patterns and can help avoid tripping the automated limits.
Use --fast-update or Smaller Batches: If you are downloading a very large number of items (e.g. stories from hundreds of profiles, or thousands of posts), consider breaking the job into smaller chunks. The --fast-update option can help by stopping when you reach media that’s already downloaded, reducing total requests on repeated runs. You can also specify --count to limit how many posts or items to fetch in one go. Fetching in batches spread over time will be gentler on the rate limit.
Rotate User Agents: Instagram might flag automated clients partly by their HTTP User-Agent string. Instaloader by default uses its own UA, but you can override it. Some developers suggest using a common browser User-Agent so the requests look more like normal web traffic. In code, you can initialize Instaloader with a custom user_agent. For multiple runs, rotating through a list of user-agent strings for each session may slightly reduce the chance of detection. (Note: This is not foolproof, but may help at the margins.)
Avoid Obvious Scraping Patterns: Try not to scrape the same data repeatedly in a short period of time. For example, do not repeatedly download the same profile or hashtag in rapid succession. Likewise, if you are scraping stories, be aware that grabbing many stories in quick succession is a known trigger. Spreading out story downloads (or limiting the number of story feeds you scrape per hour) can help.
Handle 429 Gracefully: If you get a 429 error, stop requests and wait. Instaloader’s built-in backoff will pause execution; respect it. If you’re writing a custom script and catch a TooManyRequestsException, implement a long sleep (e.g., 10–15 minutes or more) before retrying. Do not aggressively retry failed requests or you risk extending the ban.
IP and Account Rotation (Last resort): In extreme cases, if one IP address is consistently getting blocked, you might try switching to another network/IP or using a proxy with a different IP. This can bypass an IP-based throttle, but use caution: Instagram might detect account sharing or unusual IP changes. If you have multiple Instagram accounts, you could distribute your queries among them (log in with different accounts for different tasks) to stay under per-account limits. Only do this with accounts you control, and never use accounts without permission.e.
Ultimately, there is no way to completely “bypass” Instagram’s rate limits – you must work within them. The key is to slow down and behave more like a normal user. If you hit a limit, patience is the only guaranteed solution (wait until the block lifts). Trying to outsmart Instagram’s anti-scraping measures (with rapid IP rotations, etc.) often only works briefly, if at all, and can risk longer-term bans.
Ethical Considerations
Profiles with common names or aliases can be misidentified if relying on metadata alone.
Posts may contain sensitive data (location, contact info, etc.).
Ensure you have the right to view or store the content you are downloading, and respect privacy and permissions.
