Claude Fable 5: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Anthropicanthropic | Healthy | 100% | 99.88% | 99.80% | 99.82% | 3.81s | 8.51s | 56 tps | 1.0M | $10.00 / $50.00◆ (best) |
Amazon Bedrockamazon-bedrock | No data | — | — | — | — | 4.41s | 7.30s | 40 tps | 1.0M | $10.00 / $50.00◆ (best) |
Azureazure | No data | — | 99.29% | 95.14% | 96.69% | 5.02s | 5.98s | 37 tps | 1.0M | $10.00 / $50.00◆ (best) |
Claude Platform on AWSclaude-on-aws | No data | — | 100% | 84.32% | 98.30% | 3.72s | 4.89s◆ (best) | 46 tps | 1.0M | $10.00 / $50.00◆ (best) |
Googlegoogle-vertex/europe | No data | — | — | 99.30% | 99.80% | 3.56s◆ (best) | 11.5s | 73 tps◆ (best) | 1.0M | $11.00 / $55.00 |
Googlegoogle-vertex/global | No data | — | 99.40% | 100%◆ (best) | 100%◆ (best) | 5.64s | 5.64s | 57 tps | 1.0M | $10.00 / $50.00◆ (best) |
- Anthropic100%Healthy24h 99.88%
- 7d
- 99.80%
- 30d
- 99.82%
- p50
- 3.81s
- p99
- 8.51s
- Throughput
- 56 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
anthropic - No data24h —
- 7d
- —
- 30d
- —
- p50
- 4.41s
- p99
- 7.30s
- Throughput
- 40 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
amazon-bedrock - No data24h 99.29%
- 7d
- 95.14%
- 30d
- 96.69%
- p50
- 5.02s
- p99
- 5.98s
- Throughput
- 37 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
azure - No data24h 100%
- 7d
- 84.32%
- 30d
- 98.30%
- p50
- 3.72s
- p99
- 4.89s◆ (best)
- Throughput
- 46 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
claude-on-aws - No data24h —
- 7d
- 99.30%
- 30d
- 99.80%
- p50
- 3.56s◆ (best)
- p99
- 11.5s
- Throughput
- 73 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $11.00 / $55.00
google-vertex/europe - No data24h 99.40%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 5.64s
- p99
- 5.64s
- Throughput
- 57 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
google-vertex/global
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.