Two distinct layers: your Subscription Plan governs access throughput, platform features, and included monthly credits. Usage Rates meter consumption against stable virtual model IDs.
Choose the tier that fits your team's throughput requirements. All paid plans include dollar-for-dollar API usage credits every billing cycle.
For developer evaluation, prototypes, and playground experiments.
For indie developers & early-stage MVP production apps.
High-speed default for production agents & scaling apps.
High-concurrency capacity & team controls for heavy workloads.
Custom volume commitments, Bring Your Own Key (BYOK) provider credentials, organizational skill governance, and dedicated compliance support.
Transparent unit rates across virtual models. Usage draws down from your included plan credit before any additional billing occurs.
| Model Family | Default Tier | Input Rate | Output Rate | Specialization |
|---|---|---|---|---|
| dheeman-auto | MID | Target dependent | Target dependent | Heuristic meta-router (Billed at served target rate with zero surcharge) |
| dheeman-flash | LOW | $0.10 / 1M tokens | $0.40 / 1M tokens | High-volume classification, bulk text summaries, routine extractions |
| dheeman-pro | MID | $1.50 / 1M tokens | $6.00 / 1M tokens | Balanced production flagship for complex web apps, agents, synthesis |
| dheeman-reasoning | HIGH | $3.00 / 1M tokens | $12.00 / 1M tokens | Deep multi-step reasoning, mathematical proofs, complex planning |
| dheeman-coding | MID | $1.20 / 1M tokens | $4.80 / 1M tokens | Code synthesis, refactoring, test generation, AST parsing |
| dheeman-vision | MID | $1.50 / 1M tokens | $6.00 / 1M tokens | Visual document analysis, diagram interpretation, UI comprehension |
| dheeman-ocr | MID | $2.00 / 1,000 pages | — | Document layout preservation, multilingual table parsing |
| dheeman-stt | LOW | $0.15 / audio hr | — | Speech-to-text with timestamps and speaker diarization |
| dheeman-tts | LOW | $0.04 / 1,000 chars | — | Low-latency expressive neural voice synthesis |
| dheeman-image | MID | $0.04 / image | — | Photorealistic & vector generation at variable aspect ratios |
| dheeman-video | HIGH | $0.20 / video generation | — | Short-form video and motion scene rendering |
Dhee AI routes around degraded capacity automatically. If an underlying provider returns a 5xx error or latency spikes, your request is transparently re-attempted against healthy eligible routes without dropping the client connection. You are never charged for failed upstream provider attempts.
Every paid plan includes a dedicated dollar amount of usage credit. For instance, the Pro plan at $29/mo includes monthly usage credit. As your API keys invoke models, usage is deducted from this credit balance first. Once depleted, additional usage is billed at standard retail rates minus your tier discount.
Yes. In your Developer Console, you can configure both email threshold notifications and strict monthly hard budget limits per project or organization. When a hard cap is reached, non-critical requests are safely blocked with clear HTTP 429 status codes rather than generating unexpected bills.
