We tested both APIs.
Here's what we found.
8x faster search, automatic indexing, text-to-image search, and a simpler developer experience. Real benchmarks from real API calls.
Try Vecstore Free Imagga search: 2.5 seconds.
Vecstore search: ~300ms.
Vecstore's search breaks down to ~90ms for embedding and 5-8ms for the actual database search across millions of vectors. With network overhead, total round-trip is ~300ms and you get full results with image URLs and metadata in one call. Imagga's search takes 2.5s, returns only IDs, and you still need a second query to your own database to resolve image URLs.
Insert image
Vecstore
200ms
Imagga
4.3s
Search query
Vecstore
300ms
1 call, full results
Imagga
2.5s
+ separate DB query
Metadata filtering
Vecstore
Yes
Filter before search
Imagga
No
Imagga searches by tags.
Vecstore searches by actual image content.
Imagga's "visual search" works by categorizing images into tags (like border_collie.n.01 ) and then matching those tags. It's tag search, not visual similarity. Vecstore understands the actual visual content of images and finds matches by meaning.
Imagga workflow
1. Upload image to index
2. Manually train the index
3. Search (returns tag categories + IDs only)
4. Query your own database to get image URLs
Requires a separate database (e.g. Supabase) to store and resolve image URLs. New images require retraining before they're searchable.
Vecstore workflow
1. Insert image (auto-indexed)
2. Search by image or text (returns full results)
No separate database needed. Results include image URLs, metadata, and similarity scores. New images are instantly searchable.
| Feature | Imagga | Vecstore |
|---|---|---|
| How search works | Tag categorization + tag matching | Visual similarity (understands image content) |
| Search response | IDs only (need separate DB for URLs) | Full results with image URLs + metadata |
| Metadata filtering | Not supported | Built in (filter before search) |
| Text-to-image search | Not supported | Supported on all plans |
| Result labels | WordNet format (border_collie.n.01) | Developer-friendly |
| Separate database needed | Yes (to resolve image URLs) | No |
| Index management | Manual (create, train, retrain) | Fully automatic |
| Adding new images | Requires index retraining | Instantly searchable |
| Semantic text search | Not available | Built in |
Type "wind turbine." Imagga can't do this.
Text query
wind turbine
Imagga
Not supported
Vecstore
Wind turbine images found
Imagga only supports image-to-image similarity. Vecstore lets users describe what they're looking for in plain text and returns matching images from your database.
Imagga: $79/mo subscription.
Vecstore: $8 credits that never expire.
Imagga locks visual search behind their $79/mo Indie plan and unused requests expire every month. Vecstore sells credit packs that never expire, includes all features on every plan, and gives you 25 free credits on signup with no credit card.
Monthly cost by volume
5K operations
Imagga
$79/mo
expires monthly
Vecstore
$8
never expires
90% cheaper
70K operations
Imagga
$79/mo
expires monthly
Vecstore
~$66
never expires
17% cheaper
300K operations
Imagga
$349/mo
expires monthly
Vecstore
~$240
never expires
31% cheaper
| Imagga | Vecstore | |
|---|---|---|
| Free tier | 100 requests | 25 credits (no CC required) |
| Pricing model | Monthly (unused requests expire) | Credit packs (never expire) |
| Entry plan | $79/mo (70K requests) | $8 (5,000 credits) |
| Mid plan | No mid-tier option | $25 (20,000 credits) |
| Pro / Scale | $349/mo (300K requests) | ~$240 for 300K credits |
| Visual search | $79+ plans only | All plans |
| Text-to-image search | Not available | All plans |
| Face recognition | $349+ plans only | All plans |
| NSFW detection | Not available | All plans |
| Multilingual search | Not available | All plans (100+ languages) |
Imagga returns turbine.n.01 . Vecstore returns results you can use.
Imagga response (2 API calls needed)
Call 1: Search returns IDs + tag categories only
{ "categories": [{
"name": "border_collie.n.01",
"confidence": 93.4 }],
"images": [{
"id": "img_177065...",
"distance": 0.387 }] }
Call 2: Query your own database to get image URLs
SELECT image_url FROM images
WHERE save_id IN ('img_177...')
No image URLs in results. You need a separate database (Supabase, Postgres, etc.) to resolve IDs to actual images.
Vecstore response (1 API call)
Everything in one response
{ "vector_id": "abc123",
"score": 0.94,
"metadata": {
"image_url": "https://...",
"name": "Border Collie",
"category": "pets"
} }
Image URLs, custom metadata, and similarity scores. No separate database needed. One call, done.
Full comparison
| Vecstore | Imagga | |
|---|---|---|
| How search works | Visual similarity | Tag categorization + tag matching |
| Search speed | ~300ms (full results) | 2.5s (IDs only, + DB query) |
| Insert speed | ~200ms | 3.8 - 4.7s |
| Separate database needed | No | Yes (to resolve image URLs) |
| Search response | Full results + metadata + URLs | IDs only |
| Metadata filtering | Built in (pre-search filtering) | Not supported |
| Index management | Automatic | Manual (train/retrain) |
| API calls to search | 1 | 2 (search + DB query for URLs) |
| Text-to-image search | Supported | Not supported |
| Face search | All plans | $349+ plan only |
| OCR search | All plans | $79+ plan only |
| NSFW detection | 52 categories | Not available |
| Semantic text search | Supported | Not available |
| Multilingual | 100+ languages | Not available |
| Result labels | Developer-friendly | WordNet (border_collie.n.01) |
| New images searchable | Instantly | After index retrain |
| Pricing model | Credits (never expire) | Monthly subscription (expires) |
| Free tier | 25 credits, no CC | 100 requests |
| Starting price | $8 | $79/mo |
Faster search. Simpler API. Better results.
30M+ searches powered by Vecstore this year
25 Free credits. No credit card required.