Models
Available models, capabilities, and pricing for each Persly API.
Persly offers specialized models for different medical AI tasks. Each model is optimized for its specific use case.
Model Overview
| Model | API | Pricing | Description |
|---|---|---|---|
persly-chat-v1 | Chat | $0.15/request (+$0.01 when follow-up generation is invoked, +$0.003/page attachments, up to 30 pages and 40MB/request) | Standard medical Q&A with source citations |
persly-chat-pro-v1 | Chat | $0.50/request (+$0.01 when follow-up generation is invoked, +$0.01/page attachments, up to 60 pages and 40MB/request) | Advanced reasoning with tool use (web search, medical databases) |
persly-finder-v1 | Finder | $0.035–$0.07/request | AI-powered medical information search |
persly-embed-v1 | Embeddings | $0.30/1M tokens | 1536-dimensional medical text embeddings |
persly-rerank-v1 | Rerank | $15.00/1K searches | Medical document relevance ranking |
Chat Models
persly-chat-v1
Standard medical chat model. Best for general medical Q&A where speed and cost matter.
- Pricing: $0.15 per request (+$0.01 when follow-up generation is invoked, +$0.003 per attachment page, up to 30 pages and 40MB total attachments per request)
- Features: Source citations, follow-up question suggestions
- Best for: Patient-facing FAQs, general medical queries, high-volume applications
persly-chat-pro-v1
Advanced medical chat model with enhanced reasoning capabilities and access to external tools.
- Pricing: $0.50 per request (+$0.01 when follow-up generation is invoked, +$0.01 per attachment page, up to 60 pages and 40MB total attachments per request)
- Features: Everything in v1 + web search, medical database access, deeper reasoning
- Best for: Complex medical questions, differential diagnosis support, clinical decision support
Finder Model
persly-finder-v1
AI-powered search across medical information sources with optional answer generation.
| Depth | Pricing | Description |
|---|---|---|
basic | $0.035/request | Quick search, fewer sources |
advanced | $0.07/request | Deep search, more comprehensive results |
- Max results: 1–20 per request (default: 5)
- Best for: Medical information retrieval, evidence gathering, fact-checking
Embeddings Model
persly-embed-v1
Generates 1536-dimensional embedding vectors optimized for medical text similarity and retrieval.
- Pricing: $0.30 per 1M tokens
- Dimensions: 1536
- Max batch size: 100 texts per request
- Best for: Medical document search, semantic similarity, RAG pipelines
Rerank Model
persly-rerank-v1
Reranks documents by medical relevance to a query. Significantly improves search result quality.
- Pricing: $15.00 per 1K searches (1 search = up to 100 documents)
- Max documents: 1,000 per request
- Best for: Search result optimization, RAG pipeline reranking, medical literature retrieval