Widespread availability failures hit the market in the last 24 hours. GPT-5.4 Mini, GPT-5.6 Sol, GPT-5.6 Terra, GPT-5.6 Luna, Gemini 3.1 Pro (preview), and Gemini 3.1 Flash Lite all stopped responding due to credit_limit errors.
Prices
Google's Gemini family shows a range of pricing tiers. The Gemini 2.5 Flash Lite is the lowest cost option at $0.100/M for input and $0.400/M for output. The highest observed cost is for the Gemini 3.1 Pro (preview) at $2.00/M for input and $12.00/M for output.
| Model | Input Cost (/M) | Output Cost (/M) |
|---|---|---|
| Gemini 3.5 Flash | $1.50 | $9.00 |
| Gemini 2.5 Flash | $0.300 | $2.50 |
| Gemini 2.5 Flash Lite | $0.100 | $0.400 |
| Gemini 2.5 Pro | $1.25 | $10.00 |
| Gemini 3.1 Flash Lite | $0.250 | $1.50 |
| Gemini 3.1 Pro (preview) | $2.00 | $12.00 |
Speed
Median response times varied significantly, though all models measured in the last 24 hours shared an identical uptime of 33.3%. GPT-5.4 Mini recorded the fastest median response time at 831ms, followed closely by Gemini 3.1 Flash Lite at 848ms. The slowest response measured was Gemini 3.1 Pro (preview) at 3.88s.
- GPT-5.4 Mini: 831ms
- Gemini 3.1 Flash Lite: 848ms
- GPT-5.6 Terra: 1.21s
- GPT-5.6 Sol: 1.40s
- GPT-5.6 Luna: 1.51s
- Gemini 3.8 Flash: 3.26s
- Gemini 3.1 Pro (preview): 3.88s
Benchmark
Over the last 7 days, Gemini 3.8 Flash and GPT-5.6 Terra both reached 100% of available points. Gemini 3.1 Flash Lite achieved 84.4%, while GPT-5.6 Luna reached 66.7%. All four models required $0 spent to run the benchmark.
These measurements prove the current response times and the specific point of failure for the unavailable models. They do not prove the cause of the credit limit errors or the long-term stability of the providers.
