4 items
Compress the context you paste locally, then inject it into a web AI chat before sending. Send less, keep what matters. AI-Chat Compressor shrinks long chat context right in your browser, then drops only the query-relevant sentences into the chat box before you send — so you save tokens and stay under the context window. Open the side panel from the toolbar, pick the answers you want (or paste any text), type your question, and press Compress. The extension keeps the sentences that actually matter for that question and inserts them into the chat — optionally sending automatically. Why you'll like it • Save tokens and context-window space on long conversations. • Keep the relevant sentences, not whole blocks — sentence-level compression. • Everything runs locally. Your chats never leave your machine. • Works across major AI chats and 50+ interface languages. How it works 1. Reads the conversation on the current tab (or your pasted text). 2. Ranks each sentence by relevance to your question (or, with no question, by how representative it is). 3. Keeps the top sentences within a size budget you control with a Strength slider. 4. Inserts the compressed context into the chat input. Two engines (toggle in the panel) • lite — lexical BM25 + light stemming. Instant, zero download. • heavy — semantic embeddings (Model2Vec / potion, 100+ languages) that catch synonyms and paraphrase, including instruction-style queries. Downloads a small model once (~64 MB), then runs fully offline — no GPU. Private by design No accounts, no API keys, no telemetry, no tracking. Compression happens on your device. The only network request is a one-time download of the local embedding model from a public Hugging Face URL (only if you choose the heavy engine) — no personal data is sent. Where it works Works on the major AI chat websites — for example ChatGPT, Claude, and Gemini — and several others. Extra • Light/dark theme, follows your system. • "Rooms" to keep separate saved contexts. • Download the full chat history or the compressed text as a .txt.
Jul 28, 2026
rating_count is the Chrome Web Store ratings count, not a written-review count.
Media assets
Screenshots and videos on the listing.
Has promo video
Whether the listing includes at least one video.
Languages
Declared language locales.
Developer website
Listing exposes a developer website URL.
Contact email
Listing exposes a contact email.
Keyword in name
Case-insensitive substring match in the name.
Keyword in description
Case-insensitive substring match in the description.
Keyword occurrences in description
Count of case-insensitive occurrences in the description.
Category user-count percentile
Share of same-category extensions with fewer users (null if unknown).
These are transparent listing completeness / keyword signals, not a prediction of Chrome Web Store search ranking.