Qwen3.5-122B-A10B: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100%◆ (best) | 99.62% | 99.50% | 99.50% | 2.72s | 10.2s | 62 tps◆ (best) | 262K | $0.300 / $2.40 | |
| Healthy | 100%◆ (best) | 99.95% | 98.55% | 98.55% | 988ms◆ (best) | 3.20s◆ (best) | 44 tps | 262K | $0.290 / $2.40 | |
| Healthy | 100%◆ (best) | 99.81% | 99.71%◆ (best) | 99.71%◆ (best) | 1.25s | 5.43s | 32 tps | 262K | $0.400 / $3.20 | |
| Healthy | 100%◆ (best) | 99.86% | 99.24% | 99.24% | 2.27s | 7.91s | 35 tps | 262K | $0.260 / $2.08◆ (best) | |
Alibabaalibaba | Degraded | 90.91% | 94.74% | 97.87% | 97.87% | 2.65s | 10.5s | 49 tps | 262K | $0.260 / $2.08◆ (best) |
- AtlasCloud100%◆ (best)Healthy24h 99.62%
- 7d
- 99.50%
- 30d
- 99.50%
- p50
- 2.72s
- p99
- 10.2s
- Throughput
- 62 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.300 / $2.40
atlas-cloud/fp8fp8 - DeepInfra100%◆ (best)Healthy24h 99.95%
- 7d
- 98.55%
- 30d
- 98.55%
- p50
- 988ms◆ (best)
- p99
- 3.20s◆ (best)
- Throughput
- 44 tps
- Context
- 262K
- $/Mtok
- $0.290 / $2.40
deepinfra/fp4fp4 - Novita100%◆ (best)Healthy24h 99.81%
- 7d
- 99.71%◆ (best)
- 30d
- 99.71%◆ (best)
- p50
- 1.25s
- p99
- 5.43s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.400 / $3.20
novita/bf16bf16 - SiliconFlow100%◆ (best)Healthy24h 99.86%
- 7d
- 99.24%
- 30d
- 99.24%
- p50
- 2.27s
- p99
- 7.91s
- Throughput
- 35 tps
- Context
- 262K
- $/Mtok
- $0.260 / $2.08◆ (best)
siliconflow/fp8fp8 - Alibaba90.91%Degraded24h 94.74%
- 7d
- 97.87%
- 30d
- 97.87%
- p50
- 2.65s
- p99
- 10.5s
- Throughput
- 49 tps
- Context
- 262K
- $/Mtok
- $0.260 / $2.08◆ (best)
alibaba
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, averaged over the hour.
Throughput
Median tokens per second.