heeb

heeb currently supports OpenAI GPT-4o, GPT-5.5, Google AI Mode, xAI Grok-4.3, Perplexity Sonar, and Perplexity Sonar-Pro and Sonar-Reasoning-Pro. Each model consumes a different credit amount per query based on its underlying API cost, ranging from 1 credit for Grok-4.3 to 5 credits for GPT-5.5.

When you submit a POST request with your prompt and target entity, heeb returns a job_id immediately and begins processing asynchronously. You then poll a GET endpoint with that job_id to retrieve the completed structured JSON result, which typically becomes available within one to two minutes.

Each response includes a mentioned boolean flag, a sentiment classification (positive, neutral, or negative), an array of source URLs the model referenced, explicit citations the model provided in its output, and aggregated visibility and sentiment scores calculated across all queried models in the same job.

Yes. By submitting multiple programmatic jobs with different entity parameters, you can build comparative brand analytics workflows that benchmark your visibility and sentiment against competitors across every supported LLM simultaneously.

No. According to heeb's privacy policy, stored prompts and job responses are never shared with third-party AI vendors and are not used to train external machine learning models. All data is encrypted at rest and in transit using TLS 1.3, with access restricted to authorized internal personnel only.