Qwen3.6 27B: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100%◆ (best) | 99.77% | 99.80% | 99.80% | 791ms | 24.2s | 63 tps◆ (best) | 262K | $0.450 / $2.70 |
| Healthy | 100%◆ (best) | 99.93% | 99.91%◆ (best) | 99.91%◆ (best) | 554ms◆ (best) | 6.29s◆ (best) | 52 tps | 262K | $0.600 / $3.60 | |
| Healthy | 100%◆ (best) | 99.25% | 97.57% | 97.57% | 624ms | 8.34s | 32 tps | 262K | $0.320 / $3.20 | |
| Healthy | 100%◆ (best) | 96.62% | 95.42% | 95.42% | 566ms | 7.92s | 47 tps | 256K | $0.325 / $3.25 | |
| Healthy | 96.99% | 92.57% | 92.76% | 92.76% | 2.32s | 22.3s | 29 tps | 262K | $0.300 / $3.20◆ (best) | |
| Down | 96.52% | 95.08% | 72.86% | 72.86% | 4.42s | 36.5s | 20 tps | 262K | $0.300 / $2.00◆ (best) | |
Phalaphala | Down | 87.74% | 94.81% | 97.24% | 97.24% | 1.58s | 16.6s | 32 tps | 262K | $0.320 / $2.70 |
- Alibaba100%◆ (best)Healthy24h 99.77%
- 7d
- 99.80%
- 30d
- 99.80%
- p50
- 791ms
- p99
- 24.2s
- Throughput
- 63 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.450 / $2.70
alibaba - CoreWeave100%◆ (best)Healthy24h 99.93%
- 7d
- 99.91%◆ (best)
- 30d
- 99.91%◆ (best)
- p50
- 554ms◆ (best)
- p99
- 6.29s◆ (best)
- Throughput
- 52 tps
- Context
- 262K
- $/Mtok
- $0.600 / $3.60
coreweave/fp8fp8 - DeepInfra100%◆ (best)Healthy24h 99.25%
- 7d
- 97.57%
- 30d
- 97.57%
- p50
- 624ms
- p99
- 8.34s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.320 / $3.20
deepinfra/fp8fp8 - Venice100%◆ (best)Healthy24h 96.62%
- 7d
- 95.42%
- 30d
- 95.42%
- p50
- 566ms
- p99
- 7.92s
- Throughput
- 47 tps
- Context
- 256K
- $/Mtok
- $0.325 / $3.25
venice/fp8fp8 - SiliconFlow96.99%Healthy24h 92.57%
- 7d
- 92.76%
- 30d
- 92.76%
- p50
- 2.32s
- p99
- 22.3s
- Throughput
- 29 tps
- Context
- 262K
- $/Mtok
- $0.300 / $3.20◆ (best)
siliconflow/fp8fp8 - Chutes96.52%Down24h 95.08%
- 7d
- 72.86%
- 30d
- 72.86%
- p50
- 4.42s
- p99
- 36.5s
- Throughput
- 20 tps
- Context
- 262K
- $/Mtok
- $0.300 / $2.00◆ (best)
chutes/fp8fp8 - Phala87.74%Down24h 94.81%
- 7d
- 97.24%
- 30d
- 97.24%
- p50
- 1.58s
- p99
- 16.6s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.320 / $2.70
phala
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.
Throughput
Median tokens per second.