Qwen3.5-27B
qwen/qwen3.5-27b
Compare with Qwen3.5 397B A17B, Qwen3.5-35B-A3B, Qwen3.5-122B-A10B
CompareQwen3.5-27B
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100% | 97.18% | 98.16% | 98.16% | 1.07s | 7.41s | 25 tps | 262K | $0.195 / $1.56 |
| Healthy | 100% | 98.93% | 97.83% | 97.83% | 339ms | 2.37s | 51 tps | 262K | $0.260 / $2.60 | |
| Healthy | 100% | 99.71% | 99.81% | 99.81% | 1.30s | 8.37s | 12 tps | 262K | $0.300 / $2.40 | |
| Healthy | 100% | 98.84% | 98.97% | 98.97% | 1.30s | 9.19s | 10 tps | 262K | $0.250 / $2.00 | |
| Healthy | 98.48% | 99.15% | 98.98% | 98.98% | 919ms | 5.28s | 16 tps | 262K | $0.270 / $2.16 | |
Phalaphala | No data | — | 63.31% | 62.46% | 62.46% | 1.72s | 9.12s | 20 tps | 262K | $0.300 / $2.40 |
- Alibaba100%Healthy24h 97.18%
- 7d
- 98.16%
- 30d
- 98.16%
- p50
- 1.07s
- p99
- 7.41s
- Throughput
- 25 tps
- Context
- 262K
- $/Mtok
- $0.195 / $1.56
alibaba - DeepInfra100%Healthy24h 98.93%
- 7d
- 97.83%
- 30d
- 97.83%
- p50
- 339ms
- p99
- 2.37s
- Throughput
- 51 tps
- Context
- 262K
- $/Mtok
- $0.260 / $2.60
deepinfra/fp8fp8 - Novita100%Healthy24h 99.71%
- 7d
- 99.81%
- 30d
- 99.81%
- p50
- 1.30s
- p99
- 8.37s
- Throughput
- 12 tps
- Context
- 262K
- $/Mtok
- $0.300 / $2.40
novita/bf16bf16 - SiliconFlow100%Healthy24h 98.84%
- 7d
- 98.97%
- 30d
- 98.97%
- p50
- 1.30s
- p99
- 9.19s
- Throughput
- 10 tps
- Context
- 262K
- $/Mtok
- $0.250 / $2.00
siliconflow/fp8fp8 - AtlasCloud98.48%Healthy24h 99.15%
- 7d
- 98.98%
- 30d
- 98.98%
- p50
- 919ms
- p99
- 5.28s
- Throughput
- 16 tps
- Context
- 262K
- $/Mtok
- $0.270 / $2.16
atlas-cloud/fp8fp8 - No data24h 63.31%
- 7d
- 62.46%
- 30d
- 62.46%
- p50
- 1.72s
- p99
- 9.12s
- Throughput
- 20 tps
- Context
- 262K
- $/Mtok
- $0.300 / $2.40
phala
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
1 endpoint reporting no data is omitted from the charts.
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.