Public provider data covering model availability, pricing, user feedback, and risk signals. Public reports are still limited, so keep this provider under review.
Be the first to leave a review.
Be the first to ask a question.
Each 7-day reliability bar represents one day. The 24-hour view uses the same scale for the latest window: green is healthy, yellow needs review, red is weak, and gray means insufficient samples.
| Models | Group | Input/M Tokens | Output/M Tokens | Cache read/M Tokens | Cache write/M Tokens | Discount vs official | 7-day reliability | Latency | Verification rate |
|---|---|---|---|---|---|---|---|---|---|
| GPT-5 | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| GPT-5.1 | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| GPT-5.2 | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| GPT-5.2 Codex | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| GPT-5.4 | provider-data:default | -- | -- | -- | -- | -- | 0.0%2 samples | -- | No reports |
| GPT-5.4 mini | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| GPT-5.4 nano | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| Claude Haiku 4.5 (latest) | claude | -- | -- | -- | -- | -- | 26.4%265 samples | 1.9s | No reports |
| Claude Haiku 4.5 (latest) | provider-data:default | -- | -- | -- | -- | -- | --Insufficient samples | -- | No reports |
| Claude Opus 4.5 (latest) | claude | -- | -- | -- | -- | -- | 26.4%265 samples | 1.9s | No reports |
Public reviews will appear here. Sign in to share real usage experience.
Be the first to ask a question.
