Analyze text for sentiment, topics, intents, and summaries via Deepgram's NLU API, paid per-call with Tempo.
What it does
This MPP endpoint exposes Deepgram's text analysis (read-intelligence) capabilities through the Locus payment proxy. By sending a POST request with a text body, you can extract sentiment (positive/negative/neutral), generate concise summaries, identify discussed topics, and detect speaker or writer intents. The endpoint accepts a BCP-47 language code for multilingual support, defaulting to English.
The endpoint is part of a broader Deepgram service suite hosted on Locus MPP that also includes audio transcription (Nova-3, Nova-2, enhanced, base models), text-to-speech (Aura-2 voices), and a model-listing endpoint. Each call to the analyze endpoint costs 13,000 base units of pathUSD on Tempo L2, which equals $0.013 per request. Payment is handled via the MPP 402 challenge-response protocol with intent "charge" and method "tempo".
The request schema requires only a "text" field; all analysis features (sentiment, summarize, topics, intents) are toggled via boolean flags, so you pay one flat fee per call regardless of which analyses you enable. Note that the probe returned 404 on HEAD/GET — this is expected because the endpoint only accepts POST requests. The OpenAPI spec is well-defined with clear field descriptions.
Capabilities
Use cases
- —Analyzing customer feedback or reviews for sentiment and topics
- —Extracting intents from support tickets or chat transcripts
- —Generating concise summaries of long-form text content
- —Identifying discussion topics in meeting notes or articles
- —Multilingual text analysis for global content pipelines
Fit
Best for
- —Agents needing on-demand text NLU without API key management
- —Pay-per-call sentiment and topic analysis with no subscription
- —Combining with Deepgram transcription for end-to-end audio-to-insights pipelines
Not for
- —High-volume batch text processing where per-call pricing adds up
- —Real-time streaming text analysis (this is a one-shot charge endpoint)
- —Tasks requiring custom NLU model fine-tuning
Quick start
curl -X POST https://deepgram.mpp.paywithlocus.com/deepgram/analyze \
-H "Content-Type: application/json" \
-d '{
"text": "The product quality is excellent but shipping was slow.",
"sentiment": true,
"topics": true,
"summarize": true
}'Example
Request
{
"text": "The product quality is excellent but shipping was slow. I would recommend it overall.",
"topics": true,
"intents": true,
"language": "en",
"sentiment": true,
"summarize": true
}Endpoint
Quality
The OpenAPI spec is well-structured with clear request schemas and payment info, but no response schema or example responses are provided. The probe returned 404 on HEAD/GET which is expected for a POST-only endpoint, but no live 402 challenge was captured for this specific path. Crawl pages returned no useful documentation.
Warnings
- —No 402 challenge was captured for this endpoint — the probe only tried HEAD and GET, but the endpoint requires POST. The endpoint is likely live but unconfirmed by probe.
- —No response schema or example response is documented in the OpenAPI spec.
- —The currency address 0x20c000000000000000000000b9537d11c60e8b50 is assumed to be pathUSD with 6 decimals based on the Tempo method; if decimals differ, the stated price of $0.013 would be incorrect.
Citations
- —The analyze endpoint costs 13,000 base units with method tempo and intent chargehttps://deepgram.mpp.paywithlocus.com
- —Request schema requires 'text' field and supports optional sentiment, summarize, topics, intents, and language fieldshttps://deepgram.mpp.paywithlocus.com
- —Deepgram API reference documentation is available at developers.deepgram.comhttps://developers.deepgram.com/reference/deepgram-api-overview
- —LLM-readable docs are hosted at beta.paywithlocus.com/mpp/deepgram.mdhttps://beta.paywithlocus.com/mpp/deepgram.md