TVP API

Build against live model pricing and routed AI inference.

TVP keys use an OpenAI-compatible surface for model discovery, chat, responses, embeddings, audio, reranking, and video while keeping usage and key status tracked in the TVP registry.

Get API Key Manage Keys

GET/v1/models

List live models

Returns the TVP-offered live provider catalog after TVP_ALLOWED_MODELS and TVP_DISABLED_MODELS are applied.

curl https://theverificationplatform.com/v1/models \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/models/count

Count live models

Returns the count of TVP-offered models, not the raw upstream catalog size.

curl https://theverificationplatform.com/v1/models/count \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/models/{author}/{slug}/endpoints

List model endpoints

Returns provider endpoint status, pricing, throughput, latency, and supported parameters for a model.

curl https://theverificationplatform.com/v1/models/openai/gpt-4.1-mini/endpoints \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/providers

List providers

Returns provider metadata through the TVP gateway.

curl https://theverificationplatform.com/v1/providers \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/endpoints/zdr

List ZDR endpoints

Returns zero-data-retention endpoint metadata for procurement and routing checks.

curl https://theverificationplatform.com/v1/endpoints/zdr \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/activity

Read platform activity

Returns public platform activity data through the TVP gateway.

curl https://theverificationplatform.com/v1/activity \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/datasets/rankings-daily

Read daily rankings

Returns public daily ranking dataset rows for model analytics.

curl https://theverificationplatform.com/v1/datasets/rankings-daily \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

POST/v1/chat/completions

Run chat completions

Forwards OpenAI-compatible chat requests through the TVP gateway. Paid variable-output requests reserve prepaid credits before the provider call.

curl https://theverificationplatform.com/v1/chat/completions \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-4o-mini","max_tokens":128,"messages":[{"role":"user","content":"Say hello from TVP"}]}'

POST/v1/responses

Run Responses API requests

Forwards OpenAI-compatible Responses API requests for models that support the Responses shape. Include max_output_tokens for paid requests.

curl https://theverificationplatform.com/v1/responses \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-4.1-mini","max_output_tokens":128,"input":"Say hello from TVP"}'

POST/v1/messages

Run Anthropic Messages requests

Forwards Anthropic Messages-compatible requests for providers and presets that support the Messages shape.

curl https://theverificationplatform.com/v1/messages \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"anthropic/claude-sonnet-4.6","max_tokens":128,"messages":[{"role":"user","content":"Say hello from TVP"}]}'

POST/v1/presets/{slug}/chat/completions

Run preset chat completions

Forwards preset chat requests only for slugs enabled in TVP_ALLOWED_PRESETS. Include an output token limit so credits can be reserved.

curl https://theverificationplatform.com/v1/presets/support-agent/chat/completions \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"max_tokens":128,"messages":[{"role":"user","content":"Say hello from the preset"}]}'

POST/v1/presets/{slug}/messages

Run preset Messages requests

Forwards preset Anthropic Messages-compatible requests for enabled preset slugs.

curl https://theverificationplatform.com/v1/presets/support-agent/messages \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"max_tokens":128,"messages":[{"role":"user","content":"Say hello from the preset"}]}'

POST/v1/presets/{slug}/responses

Run preset Responses requests

Forwards preset Responses API requests for enabled preset slugs.

curl https://theverificationplatform.com/v1/presets/support-agent/responses \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"max_output_tokens":128,"input":"Say hello from the preset"}'

POST/v1/embeddings

Generate embeddings

Forwards OpenAI-compatible embedding requests for embedding-capable models when the upstream provider supports the model.

curl https://theverificationplatform.com/v1/embeddings \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen/qwen3-embedding-8b","input":"TVP model procurement"}'

GET/v1/embeddings/models

List embedding models

Returns embedding-specific model discovery data filtered to models TVP offers.

curl https://theverificationplatform.com/v1/embeddings/models \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

POST/v1/rerank

Rerank documents

Forwards rerank requests for rerank-capable models and records returned usage or generation metadata.

curl https://theverificationplatform.com/v1/rerank \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"cohere/rerank-v3.5","query":"capital of France","documents":["Paris is in France","Berlin is in Germany"]}'

POST/v1/audio/speech

Generate speech

Forwards text-to-speech requests and streams the returned audio bytes to the client.

curl https://theverificationplatform.com/v1/audio/speech \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/gpt-4o-mini-tts-2025-12-15","input":"Hello from TVP","voice":"alloy","response_format":"mp3"}' \
  --output speech.mp3

POST/v1/audio/transcriptions

Transcribe speech

Forwards speech-to-text requests with base64-encoded audio and records returned usage.

curl https://theverificationplatform.com/v1/audio/transcriptions \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"openai/whisper-1","input_audio":{"data":"BASE64_AUDIO","format":"wav"}}'

GET/v1/videos/models

List video models

Returns video-generation model metadata filtered to models TVP offers.

curl https://theverificationplatform.com/v1/videos/models \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

POST/v1/videos

Submit video generation

Starts an owner-bound asynchronous video generation job with a prepaid reservation. Poll the returned job until metering completes, then download the content.

curl https://theverificationplatform.com/v1/videos \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"google/veo-3.1-fast","prompt":"A product demo shot on a clean studio desk"}'

GET/v1/videos/{jobId}

Poll video generation

Returns current video job status only to the TVP key that created the job and meters completed jobs exactly once when usage is available.

curl https://theverificationplatform.com/v1/videos/JOB_ID \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/videos/{jobId}/content

Download video content

Streams generated video bytes only after ownership passes and the completed job has been reconciled for metering.

curl https://theverificationplatform.com/v1/videos/JOB_ID/content \
  -H "Authorization: Bearer tvp_live_YOUR_KEY" \
  --output video.mp4

GET/v1/generation

Read generation metadata

Returns generation cost and token metadata only when the generation belongs to the authenticated TVP key.

curl "https://theverificationplatform.com/v1/generation?id=gen_123" \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/generation/content

Read generation content

Returns generation content only after the same owner check used for generation metadata passes.

curl "https://theverificationplatform.com/v1/generation/content?id=gen_123" \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

GET/v1/credits

Check TVP credits

Returns TVP credit totals and usage for the authenticated TVP key. This reflects TVP billing, not upstream provider account credits.

curl https://theverificationplatform.com/v1/credits \
  -H "Authorization: Bearer tvp_live_YOUR_KEY"

Limits and model support

Key status enforcement

Revoked or paused keys are blocked before any upstream provider call is made.

Request limits

Each key has a monthly request cap and a short-window rate limit enforced server-side.

Provider availability

Final model availability depends on upstream endpoint health, provider permissions, and supported modalities.

Usage tracking

TVP records endpoint, status, model, and token usage when returned by the provider response.

Prepaid reservations

Paid requests reserve estimated credits before the upstream call. Successful metered usage settles to actual cost and unused reserve is released.

Output bounds

Variable-output paid routes require max_tokens, max_completion_tokens, or max_output_tokens. Use tvp_max_cost_micros for custom bounded workflows.

Streaming usage

Paid streaming requests must set stream_options.include_usage=true so TVP can settle the reservation from provider usage.

Generation privacy

Generation metadata and content routes require durable TVP ledger ownership proof before any upstream generation lookup is made.

Preset allowlist

Preset routes are disabled unless the slug is listed in TVP_ALLOWED_PRESETS.

Platform-managed controls

Upstream BYOK, guardrails, workspaces, organization, observability, provider key management, user model lists, and upstream credit purchases are managed by TVP rather than exposed as direct customer pass-through endpoints.