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 | 99.08% | 99.75% | 99.82% | 99.82% | 3.28s◆ (best) | 13.8s | 51 tps | 1.0M | $10.00 / $50.00◆ (best) |
Amazon Bedrockamazon-bedrock | No data | — | — | — | — | 5.33s | 7.18s | 39 tps | 1.0M | $10.00 / $50.00◆ (best) |
Azureazure | No data | — | 99.83% | 95.12% | 96.91% | 5.05s | 6.64s | 51 tps | 1.0M | $10.00 / $50.00◆ (best) |
Claude Platform on AWSclaude-on-aws | No data | — | 100% | 84.32% | 98.46% | 5.20s | 29.2s | 45 tps | 1.0M | $10.00 / $50.00◆ (best) |
Googlegoogle-vertex/europe | No data | — | 99.68% | 99.30% | 99.80% | — | — | — | 1.0M | $11.00 / $55.00 |
Googlegoogle-vertex/global | No data | — | 99.87% | 100%◆ (best) | 100%◆ (best) | 4.45s | 5.34s◆ (best) | 55 tps◆ (best) | 1.0M | $10.00 / $50.00◆ (best) |
- Anthropic99.08%Healthy24h 99.75%
- 7d
- 99.82%
- 30d
- 99.82%
- p50
- 3.28s◆ (best)
- p99
- 13.8s
- Throughput
- 51 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
anthropic - No data24h —
- 7d
- —
- 30d
- —
- p50
- 5.33s
- p99
- 7.18s
- Throughput
- 39 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
amazon-bedrock - No data24h 99.83%
- 7d
- 95.12%
- 30d
- 96.91%
- p50
- 5.05s
- p99
- 6.64s
- Throughput
- 51 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
azure - No data24h 100%
- 7d
- 84.32%
- 30d
- 98.46%
- p50
- 5.20s
- p99
- 29.2s
- Throughput
- 45 tps
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
claude-on-aws - No data24h 99.68%
- 7d
- 99.30%
- 30d
- 99.80%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 1.0M
- $/Mtok
- $11.00 / $55.00
google-vertex/europe - No data24h 99.87%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 4.45s
- p99
- 5.34s◆ (best)
- Throughput
- 55 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $10.00 / $50.00◆ (best)
google-vertex/global
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.
Throughput
Median tokens per second.