Read a split Instagram JSON conversation without uploading it

If your Instagram conversation is spread across message_1.json and message_2.json, opening one file can leave part of the conversation out. Garbled emoji are a different problem: the text may need decoding, rather than a new download.

The Tera Chat Instagram viewer handles these files in your browser. Open the website, choose the export and read the conversation. You do not need to install the Mac app, write Python, save an HTML program or give an assistant your DMs.

Try a complete example before choosing your own messages

We created a small fictional archive to make the result checkable. It contains no real messages or Instagram accounts. Download the example ZIP, open the viewer, select Choose your chat export, and choose the ZIP. Select jose and mira, then Open.

You should see eight messages, starting with “Coffee after the walk? ☕” and ending with “On my way!” The name José, the café accent and the emoji should be readable. The archive also lists separate example, a different conversation containing two messages. Opening one conversation does not add the other conversation’s messages to it.

What you open	Expected result in this fictional example
Only message_1.json	Four messages; the earlier part is missing
Only message_2.json	Four earlier messages
Both parts of José and Mira’s conversation	Eight messages, sorted by timestamp
Example ZIP, then jose and mira	The same eight messages
Example ZIP, then separate example	Two messages from a separate conversation

The first part also contains a “Liked a message” entry. That entry is excluded from the message count. Eight messages does not mean eight raw JSON records.

To try the loose-file route, download message_1.json and message_2.json, then select both files together in the file chooser. We verified these sample results on 2 November 2026 using the same parser served by the live tools. The example demonstrates supported cases; it is not a guarantee for every archive size or export format.

Keep all parts of one conversation together

In a JSON export, look for a conversation folder under messages/inbox, messages/message_requests or messages/archived_threads. A longer conversation can have several numbered message_N.json files. Choose all of that conversation’s parts, rather than assuming the first filename means the complete history.

Tera Chat can read those parts inside one ZIP and offers a conversation picker. If you choose loose JSON files instead, it combines the selected files into one conversation. Do not select files from different conversation folders together. It also does not deduplicate overlapping copies of the same export: importing the same messages twice can inflate counts.

If Instagram gave you several separate ZIP archives, keep the original copies. For a conversation split between those archives, extract its JSON parts and choose those parts together. Do not assume selecting several ZIPs will combine the same conversation across them.

Why emoji and accented names can look broken

One common export encoding issue makes José appear as José, with similar garbling in emoji. The viewer attempts to reinterpret those byte values as UTF-8. Our sample deliberately includes this encoding issue so you can compare the JSON with the readable result.

This repairs a supported encoding pattern. It cannot reconstruct a character that has already been replaced by a question mark or removed from the file. Keep the original JSON untouched and compare a few names and messages with the app before relying on the result.

Read, search and save the result

The Instagram viewer opens the chat as message bubbles. Choose which person is you to place that person’s messages on the right. Use Download for Excel (CSV) for one row per message, or Save as PDF to open your browser’s print dialog. PDF controls differ by browser and device.

For comparison, here is the expected eight-message CSV. Its timestamps were generated in the Asia/Kolkata time zone; your display and CSV use your device’s local time zone. The messages and their order should agree even when the clock times differ.

Search filters the saved output too. Clear the search box before exporting the whole conversation. The PDF and CSV contain text and attachment placeholders, not the actual photos, videos or voice recordings. To see counts and reply-time statistics instead, open the same file in the Instagram chat analyzer.

What local processing means here

Choosing the file allows this page to read it on your device. The chat content is not sent to Tera Chat or an AI model for analysis. Loading the website still makes ordinary web requests; that is separate from uploading a conversation. After the page finishes loading, you can disconnect before selecting the fictional sample to check that its processing works without a connection. That demonstration is useful evidence, not a complete security audit.

On a phone, open the website and select a saved file from Files or Downloads. This avoids having to run a downloaded HTML program in a mobile browser. Very large archives can exceed a phone’s available memory; use a computer or choose the extracted parts of just the conversation you need.

When to choose another route

If you downloaded Instagram’s HTML export and only want to read it, try opening those HTML files directly in a browser. Tera Chat’s tool requires JSON. If you want an Instagram-style viewer with exported photos and audio, the open-source DevFaraaz Instagram DM Viewer documents JSON and HTML support, media display and local processing. Its developer also documents memory limits on mobile. We reviewed that documentation, not a security audit or a device-by-device test.

Use Tera Chat for a local, searchable text view, CSV/PDF output or numerical chat statistics from the JSON files you already have. Check the export’s date range and included parts first. A readable conversation is not proof that missing, deleted or unsent messages have been recovered.