Claude Fable 5.2: 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%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 4.87s | 44.0s | 76 tps | 1.0M | $8.00 / $40.00◆ (best) |
Anthropicanthropic/fast | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 1.93s◆ (best) | 12.1s◆ (best) | 252 tps◆ (best) | 1.0M | $16.00 / $80.00 |
Azureazure | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 5.55s | 52.7s | 71 tps | 1.0M | $8.00 / $40.00◆ (best) |
Googlegoogle-vertex/global | Healthy | 100%◆ (best) | 100% | 100%◆ (best) | 100%◆ (best) | 4.78s | 81.3s | 79 tps | 1.0M | $8.00 / $40.00◆ (best) |
Amazon Bedrockamazon-bedrock | No data | — | — | — | — | 5.59s | 96.4s | 55 tps | 1.0M | $8.00 / $40.00◆ (best) |
- Anthropic100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 4.87s
- p99
- 44.0s
- Throughput
- 76 tps
- Context
- 1.0M
- $/Mtok
- $8.00 / $40.00◆ (best)
anthropic - Anthropic100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 1.93s◆ (best)
- p99
- 12.1s◆ (best)
- Throughput
- 252 tps◆ (best)
- Context
- 1.0M
- $/Mtok
- $16.00 / $80.00
anthropic/fast - Azure100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 5.55s
- p99
- 52.7s
- Throughput
- 71 tps
- Context
- 1.0M
- $/Mtok
- $8.00 / $40.00◆ (best)
azure - Google100%◆ (best)Healthy24h 100%
- 7d
- 100%◆ (best)
- 30d
- 100%◆ (best)
- p50
- 4.78s
- p99
- 81.3s
- Throughput
- 79 tps
- Context
- 1.0M
- $/Mtok
- $8.00 / $40.00◆ (best)
google-vertex/global - No data24h —
- 7d
- —
- 30d
- —
- p50
- 5.59s
- p99
- 96.4s
- Throughput
- 55 tps
- Context
- 1.0M
- $/Mtok
- $8.00 / $40.00◆ (best)
amazon-bedrock
Measured 4m 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.