Native
the IR surface (M2): generate, embed, rerank, count_tokens, dry_run
/v1/aiml/embedAuthorization
bearer In: header
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
application/json
application/json
curl -X POST "https://example.com/v1/aiml/embed" \ -H "Content-Type: application/json" \ -d '{ "model": "voyage/voyage-3", "input": [ "what does the ledger do?" ], "input_type": "query" }'{ "object": "list", "id": "01J8REQ000000000000000AB01", "model": "voyage/voyage-3", "data": [ { "index": 0, "embedding": [ 0.0123, -0.0456 ], "dimensions": 1024 } ], "usage": { "input_tokens": 7, "total_tokens": 7, "source": "provider" }, "resolved": { "provider": "voyage", "endpoint_id": "ep_voyage_voyage-3_us" }, "cost": { "total_micro_usd": 1 }}/v1/aiml/rerankAuthorization
bearer In: header
Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
Response Body
application/json
application/json
application/json
curl -X POST "https://example.com/v1/aiml/rerank" \ -H "Content-Type: application/json" \ -d '{ "model": "cohere/rerank-3.5", "query": "how do credits expire?", "documents": [ "grants expire after 14 days", "the gateway streams tokens" ], "top_n": 1, "return_documents": true }'{ "object": "rerank", "id": "01J8REQ000000000000000AB02", "model": "cohere/rerank-3.5", "results": [ { "index": 0, "relevance_score": 0.91, "document": "grants expire after 14 days" } ], "usage": { "input_tokens": 18, "search_units": 1, "source": "provider" }, "cost": { "total_micro_usd": 9 }}/v1/aiml/generateAuthorization
bearer In: header
Header Parameters
API version (YYYY-MM-DD); defaults to the key's pinned version
datereplay-safe retries of billed calls (part2a §7.1)
length <= 128Request Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
ai.ml canonical generate request (docs/plan part2a §6.2). aiml-version 2026-09-09.
Response Body
application/json
application/json
application/json
application/json
curl -X POST "https://example.com/v1/aiml/generate" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/gpt-5-mini", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Summarise the attached invoice in one line." } ] } ], "generation": { "max_tokens": 256, "temperature": 0.2 }, "route": { "sort": "price", "fallbacks": true }, "metadata": { "app": "chatbot", "user_id": "u_42" } }'{ "id": "01J9ZK2K4Q7Q3M3QX1V0F1GATE", "model": "openai/gpt-5-mini", "candidates": [ { "index": 0, "content": [ { "type": "text", "text": "Invoice AIML/26-27/000001: ₹1,180.00 for prepaid credits, GST included." } ], "stop": { "reason": "end_turn" } } ], "usage": { "input_tokens": 812, "output_tokens": 143, "cache_read_tokens": 0, "cache_write_tokens": 0, "reasoning_tokens": 0 }, "cost": { "total_micro_usd": 489, "lines": [ { "kind": "input", "units": 812, "unit_price_nano": 250, "micro_usd": 203 }, { "kind": "output", "units": 143, "unit_price_nano": 2000, "micro_usd": 286 } ], "price_version": "pv_01J9ZK2K4Q7Q3M3QX1V0F1PV01" }, "provider_meta": { "provider": "openai", "endpoint_id": "ep_openai_gpt-5-mini_us" }, "warnings": []}/v1/aiml/count_tokensAuthorization
bearer In: header
Header Parameters
API version (YYYY-MM-DD); defaults to the key's pinned version
dateRequest Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
ai.ml canonical generate request (docs/plan part2a §6.2). aiml-version 2026-09-09.
Response Body
application/json
application/json
curl -X POST "https://example.com/v1/aiml/count_tokens" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/gpt-5-mini", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "hi" } ] } ] }'{ "input_tokens": 812, "source": "exact", "tokenizer": "o200k_base"}/v1/aiml/dry_runAuthorization
bearer In: header
Header Parameters
API version (YYYY-MM-DD); defaults to the key's pinned version
dateRequest Body
application/json
TypeScript Definitions
Use the request body type in TypeScript.
ai.ml canonical generate request (docs/plan part2a §6.2). aiml-version 2026-09-09.
Response Body
application/json
application/json
application/json
curl -X POST "https://example.com/v1/aiml/dry_run" \ -H "Content-Type: application/json" \ -d '{ "model": "openai/gpt-5-mini", "messages": [ { "role": "user", "content": [ { "type": "text", "text": "Summarise the attached invoice in one line." } ] } ], "generation": { "max_tokens": 256, "temperature": 0.2 }, "route": { "sort": "price", "fallbacks": true }, "metadata": { "app": "chatbot", "user_id": "u_42" } }'{ "plan": [ { "endpoint_id": "ep_openai_gpt-5-mini_us", "provider": "openai", "score": 0.97 } ], "warnings": [], "reserve_micro": 2600, "price_version": "pv_01J9ZK2K4Q7Q3M3QX1V0F1PV01"}