provider
api.commandcode.ai
Provider overview built from anonymous public model verification runs, including test volume, match rate, popular models, and recent samples.
How to read this page
The overall score is better suited to comparisons than raw match rate: it limits tiny perfect samples and rewards leading GPT / Claude coverage, verified model breadth, sufficient evidence, and recent activity. Still combine it with recent samples, verdicts, and latency.
Evidence scope and observation dates
- Matches / diagnostic samples
- 5 / 5
- Diagnostic samples in the last 30 days
- 2
- Latest recorded test
The match-rate denominator is diagnostic samples with a verdict and model label, rather than all public tests. These observations are not independent-user counts, a random sample or coverage of every request served by this provider.
The table preserves version information present in the requested model ID and claimed model label. Labels alone do not independently prove the upstream version; one mismatch is insufficient to establish fraud.
Recent samples
| Requested model ID | Claimed model | Verdict | Latency | Completed (UTC) |
|---|---|---|---|---|
| deepseek/deepseek-v4.1-flash | deepseek-v4.1-flash | match | 4.5s | |
| deepseek/deepseek-v4-flash | deepseek-v4.1-flash | match | 5.5s | |
| deepseek/deepseek-v4-flash | deepseek-v4-flash | match | 3.7s | |
| deepseek/deepseek-v4-flash | deepseek-v4-flash | match | 3.6s | |
| gpt-5.6-sol | gpt-5.6-sol | match | 3.4s |