GPT-6.1 Sol: 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/eu | Healthy | 100%◆ (best) | 99.96% | 99.99% | 99.99% | 2.56s | 10.7s | 46 tps | 1.1M | $2.20 / $11.00 |
Azureazure/us | Healthy | 100%◆ (best) | 99.36% | 99.02% | 99.02% | 1.35s◆ (best) | 8.41s◆ (best) | 27 tps | 1.1M | $2.20 / $11.00 |
OpenAIopenai/fast | Healthy | 100%◆ (best) | 99.98% | 100%◆ (best) | 100%◆ (best) | 1.77s | 64.0s | 27 tps | 1.1M | $4.00 / $20.00 |
OpenAIopenai/flex | Healthy | 99.98% | 99.94% | 99.97% | 99.97% | 5.41s | 67.0s | 48 tps◆ (best) | 1.1M | $1.00 / $5.00◆ (best) |
Azureazure | Healthy | 99.54% | 99.94% | 99.88% | 99.88% | 5.20s | 39.7s | 40 tps | 1.1M | $2.00 / $10.00 |
OpenAIopenai | Healthy | 97.98% | 99.85% | 99.82% | 99.82% | 4.80s | 63.2s | 23 tps | 1.1M | $2.00 / $10.00 |
- Azure100%◆ (best)Healthy24h 99.96%
- 7d
- 99.99%
- 30d
- 99.99%
- p50
- 2.56s
- p99
- 10.7s
- Throughput
- 46 tps
- Context
- 1.1M
- $/Mtok
- $2.20 / $11.00
azure/eu - Azure100%◆ (best)Healthy24h 99.36%
- 7d
- 99.02%
- 30d
- 99.02%
- p50
- 1.35s◆ (best)
- p99
- 8.41s◆ (best)
- Throughput
- 27 tps
- Context
- 1.1M
- $/Mtok
- $2.20 / $11.00
azure/us - OpenAI100%◆ (best)Healthy24h 99.98%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 1.77s
- p99
- 64.0s
- Throughput
- 27 tps
- Context
- 1.1M
- $/Mtok
- $4.00 / $20.00
openai/fast - OpenAI99.98%Healthy24h 99.94%
- 7d
- 99.97%
- 30d
- 99.97%
- p50
- 5.41s
- p99
- 67.0s
- Throughput
- 48 tps◆ (best)
- Context
- 1.1M
- $/Mtok
- $1.00 / $5.00◆ (best)
openai/flex - Azure99.54%Healthy24h 99.94%
- 7d
- 99.88%
- 30d
- 99.88%
- p50
- 5.20s
- p99
- 39.7s
- Throughput
- 40 tps
- Context
- 1.1M
- $/Mtok
- $2.00 / $10.00
azure - OpenAI97.98%Healthy24h 99.85%
- 7d
- 99.82%
- 30d
- 99.82%
- p50
- 4.80s
- p99
- 63.2s
- Throughput
- 23 tps
- Context
- 1.1M
- $/Mtok
- $2.00 / $10.00
openai
Measured 2m 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.