Similarity & Duplicate Detection API
## Fast Text & Image Similarity API Compare texts, images, or cross-modal (text vs image) using state-of-the-art ML embeddings. No LLM costs, no per-token pricing — just fast, affordable similarity scoring. ### Models Available - **minilm** (default) — Fast English text similarity (~10ms) - **mpnet** — Higher quality English text (~20ms) - **multilingual** — 50+ languages including Italian,…
1 subscribers
9.2/10 popularity
13342 ms avg latency
49% success rate
9 endpoints
The in-depth APIMemo review for this API hasn't been published yet —
the data below comes straight from the public marketplace listing.
Similarity & Duplicate Detection API endpoints
| Method | Endpoint | Description |
|---|---|---|
| System | ||
| GET |
usage_stats /usage |
Check your API key usage stats for the current month. Returns plan info, request count, and per-endpoint breakdown. |
| GET |
list_plans /plans |
List available plans with their monthly request limits and pricing. |
| GET |
health_check /health |
Check if the API is running and which models are loaded. |
| Similarity | ||
| POST |
cross_modal_similarity /similarity/cross |
Compare text against an image and get a similarity score (0-1). Uses CLIP to embed both in the same space. Great for image search by text query. |
| POST |
text_similarity /similarity/text |
Compare two texts and get a similarity score (0-1). Supports models: minilm (fast), mpnet (better quality), multilingual (50+ languages). Long texts are handled automatically via… |
| POST |
image_similarity /similarity/image |
Compare two images by URL and get a visual similarity score (0-1). Uses CLIP ViT-B/32 model. Images must be publicly accessible URLs, max 10MB each. |
| Batch | ||
| POST |
batch_text_similarity /similarity/batch/text |
Compare a query text against a list of candidates. Returns the top_k most similar, ranked by score. Ideal for finding duplicates or similar items from a list. |
| POST |
batch_image_similarity /similarity/batch/image |
Compare a query image against a list of candidate images. Returns the top_k most visually similar. Max 20 candidate images per request. |
| Embeddings | ||
| POST |
get_embeddings /embeddings |
Get raw embedding vectors for texts or images. Use these to store in your own vector database for custom similarity search. Supports up to 32 inputs per request. |
Similarity & Duplicate Detection API pricing
| Plan | Price | Rate limit | Quotas |
|---|---|---|---|
| BASIC | Free | — |
|
| PRO | $5 / month | — |
|
| ULTRA | $19 / month | — |
|
| MEGA | $50 / month | — |
|
