Claude Sonnet 4.6: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Amazon Bedrockamazon-bedrock/global | Healthy | 100%◆ (best) | 99.96% | 99.95% | 99.95% | 1.29s | 5.86s | 40 tps | 1.0M | $3.00 / $15.00◆ (best) |
Anthropicanthropic | Healthy | 100%◆ (best) | 99.90% | 99.85% | 99.85% | 1.63s | 11.0s | 34 tps | 1.0M | $3.00 / $15.00◆ (best) |
Googlegoogle-vertex/global | Healthy | 100%◆ (best) | 99.97% | 99.96%◆ (best) | 99.96%◆ (best) | 1.47s | 5.70s | 37 tps | 1.0M | $3.00 / $15.00◆ (best) |
Claude Platform on AWSclaude-on-aws | Healthy | 99.98% | 99.91% | 99.90% | 99.90% | 1.08s | 3.66s | 38 tps | 1.0M | $3.00 / $15.00◆ (best) |
Amazon Bedrockamazon-bedrock/eu-west-1 | No data | — | — | — | — | — | — | — | 1.0M | $3.30 / $16.50 |
Amazon Bedrockamazon-bedrock/us | No data | — | — | — | — | — | — | — | 1.0M | $3.30 / $16.50 |
Azureazure/global | No data | — | 100% | 99.78% | 99.78% | 1.55s | 43.4s | 46 tps | 1.0M | $3.00 / $15.00◆ (best) |
Googlegoogle-vertex/europe | No data | — | 99.83% | 99.52% | 99.52% | 1.06s | 8.95s | 50 tps◆ (best) | 1.0M | $3.30 / $16.50 |
Googlegoogle-vertex/us-east5 | No data | — | 100% | — | — | 959ms◆ (best) | 1.06s◆ (best) | 18 tps | 1.0M | $3.30 / $16.50 |
- Amazon Bedrock100%◆ (best)Healthy24h 99.96%
- 7d
- 99.95%
- 30d
- 99.95%
- p50
- 1.29s
- p99
- 5.86s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00◆ (best)
amazon-bedrock/global - Anthropic100%◆ (best)Healthy24h 99.90%
- 7d
- 99.85%
- 30d
- 99.85%
- p50
- 1.63s
- p99
- 11.0s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00◆ (best)
anthropic - Google100%◆ (best)Healthy24h 99.97%
- 7d
- 99.96%◆ (best)
- 30d
- 99.96%◆ (best)
- p50
- 1.47s
- p99
- 5.70s
- Throughput
- 37 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00◆ (best)
google-vertex/global - Claude Platform on AWS99.98%Healthy24h 99.91%
- 7d
- 99.90%
- 30d
- 99.90%
- p50
- 1.08s
- p99
- 3.66s
- Throughput
- 38 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00◆ (best)
claude-on-aws - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
amazon-bedrock/eu-west-1 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
amazon-bedrock/us - No data24h 100%
- 7d
- 99.78%
- 30d
- 99.78%
- p50
- 1.55s
- p99
- 43.4s
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $3.00 / $15.00◆ (best)
azure/global - No data24h 99.83%
- 7d
- 99.52%
- 30d
- 99.52%
- p50
- 1.06s
- p99
- 8.95s
- Throughput
- 50 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
google-vertex/europe - No data24h 100%
- 7d
- —
- 30d
- —
- p50
- 959ms◆ (best)
- p99
- 1.06s◆ (best)
- Throughput
- 18 tps
- Context
- 1.0M
- $/Mtok
- $3.30 / $16.50
google-vertex/us-east5
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.
Throughput
Median tokens per second.