Model Catalog
Observed model usage joined to verified rate-card entries.
Loading…
Loading…
Observed model usage joined to verified rate-card entries.
Loading…
4
Models observed
Current dataset
4
Providers observed
Current dataset
2
Verified rate entries
Current card
2
Pricing gaps
Require verification
Requires benchmark or equivalence evidence
The telemetry API measures usage, latency, failures, and cost. It does not manufacture a quality score from those signals. Connect the equivalence proof pipeline before plotting a price-versus-quality frontier.
No proof-backed recommendations are published by the API
Model fit and projected savings stay empty until a candidate has passed the workload's declared equivalence criteria.
| Provider | Model | Input rate | Output rate | Modality | Observed latency | Active workloads | Pricing status |
|---|---|---|---|---|---|---|---|
| Anthropic | anthropic/claude-sonnet-4.6 | Unpriced | Unpriced | Text | 52.6s | 5 | Evaluating |
| Openai | openai/gpt-5-mini | Unpriced | Unpriced | Text | 48.3s | 5 | Evaluating |
| Openrouter | deepseek/deepseek-chat-v3.1 | Unpriced | Unpriced | Text | 54.2s | 5 |
2 vendor/model pairs covering 5 calls have no verified price
| Provider | Model | Modality | Calls | Units |
|---|---|---|---|---|
| Openai | openai/gpt-5-mini | Text | 3 | 41.9K |
| Anthropic | anthropic/claude-sonnet-4.6 | Text | 2 | 13.1K |
| Evaluating |
| google/gemini-2.5-flash-lite | Unpriced | Unpriced | Text | 58.1s | 4 | Evaluating |