GPT-6 Luna Pro: providers compared
Every endpoint serving this model, side by side. ◆ marks the best value in each column — per model, never fleet-wide.
Current state
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Azureazure | Healthy | 100%◆ (best) | 99.93% | 99.99% | 99.36% | 7.37s | 32.4s | 100 tps | 1.1M | $0.100 / $0.500 |
OpenAIopenai | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 99.90% | 8.54s | 59.6s | 97 tps | 1.1M | $0.100 / $0.500 |
Azureazure/eu | No data | — | 100% | 99.97% | 99.99%◆ (best) | — | — | — | 1.1M | $0.110 / $0.550 |
Azureazure/us | No data | — | 100% | 100%◆ (best) | 99.86% | — | — | — | 1.1M | $0.110 / $0.550 |
OpenAIopenai/fast | No data | — | 100% | 100%◆ (best) | 99.93% | 5.20s◆ (best) | 24.8s◆ (best) | 156 tps◆ (best) | 1.1M | $0.200 / $1.00 |
OpenAIopenai/flex | No data | — | 100% | 80.10% | 90.52% | 22.4s | 52.8s | 57 tps | 1.1M | $0.050 / $0.250◆ (best) |
- Azure100%◆ (best)Healthy24h 99.93%
- 7d
- 99.99%
- 30d
- 99.36%
- p50
- 7.37s
- p99
- 32.4s
- Throughput
- 100 tps
- Context
- 1.1M
- $/Mtok
- $0.100 / $0.500
azure - OpenAI100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.90%
- p50
- 8.54s
- p99
- 59.6s
- Throughput
- 97 tps
- Context
- 1.1M
- $/Mtok
- $0.100 / $0.500
openai - No data24h 100%
- 7d
- 99.97%
- 30d
- 99.99%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.1M
- $/Mtok
- $0.110 / $0.550
azure/eu - No data24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.86%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.1M
- $/Mtok
- $0.110 / $0.550
azure/us - No data24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.93%
- p50
- 5.20s◆ (best)
- p99
- 24.8s◆ (best)
- Throughput
- 156 tps◆ (best)
- Context
- 1.1M
- $/Mtok
- $0.200 / $1.00
openai/fast - No data24h 100%
- 7d
- 80.10%
- 30d
- 90.52%
- p50
- 22.4s
- p99
- 52.8s
- Throughput
- 57 tps
- Context
- 1.1M
- $/Mtok
- $0.050 / $0.250◆ (best)
openai/flex
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows. ◆ marks the best value in each column among these endpoints.
Price vs latency
One dot per endpoint at the latest round · further left = faster, lower = cheaper · hover a dot for its tag
History
Availability
Axis starts at 90% so small differences stay visible; it expands to 0 when an endpoint falls below.
Time to first token
Logarithmic axis — the fleet spans two orders of magnitude. P50 is OpenRouter's rolling 30-minute window, averaged over the hour.
Throughput
Median tokens per second.