GPT-5.6 Terra 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 | No data | — | 100% | 100%◆ (best) | 99.99% | 2.43s | 35.2s | 32 tps | 1.1M | $2.00 / $12.00 |
Azureazure/eu | No data | — | — | — | 100%◆ (best) | — | — | — | 1.1M | $2.20 / $13.20 |
OpenAIopenai | No data | — | 99.99% | 99.87% | 99.87% | 6.89s | 35.2s | 124 tps◆ (best) | 1.1M | $2.00 / $12.00 |
OpenAIopenai/fast | No data | — | 100% | — | 100%◆ (best) | 1.77s◆ (best) | 1.78s◆ (best) | 33 tps | 1.1M | $4.00 / $24.00 |
OpenAIopenai/flex | No data | — | — | — | 100%◆ (best) | — | — | — | 1.1M | $1.00 / $6.00◆ (best) |
OpenAIopenai/priority | No data | — | — | — | — | — | — | — | 1.1M | $4.00 / $24.00 |
- No data24h 100%
- 7d
- 100%◆ (best)
- 30d
- 99.99%
- p50
- 2.43s
- p99
- 35.2s
- Throughput
- 32 tps
- Context
- 1.1M
- $/Mtok
- $2.00 / $12.00
azure - No data24h —
- 7d
- —
- 30d
- 100%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.1M
- $/Mtok
- $2.20 / $13.20
azure/eu - No data24h 99.99%
- 7d
- 99.87%
- 30d
- 99.87%
- p50
- 6.89s
- p99
- 35.2s
- Throughput
- 124 tps◆ (best)
- Context
- 1.1M
- $/Mtok
- $2.00 / $12.00
openai - No data24h 100%
- 7d
- —
- 30d
- 100%◆ (best)
- p50
- 1.77s◆ (best)
- p99
- 1.78s◆ (best)
- Throughput
- 33 tps
- Context
- 1.1M
- $/Mtok
- $4.00 / $24.00
openai/fast - No data24h —
- 7d
- —
- 30d
- 100%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.1M
- $/Mtok
- $1.00 / $6.00◆ (best)
openai/flex - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.1M
- $/Mtok
- $4.00 / $24.00
openai/priority
Measured 5m 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.