Qwen3.5 397B A17B: 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) | 99.85% | 99.53%◆ (best) | 99.75%◆ (best) | 962ms | 5.65s | 101 tps◆ (best) | 262K | $0.390 / $2.34◆ (best) |
| Healthy | 100%◆ (best) | 99.72% | 99.08% | 97.58% | 993ms | 2.28s | 53 tps | 262K | $0.450 / $3.00 | |
Novitanovita | Healthy | 100%◆ (best) | 99.39% | 98.21% | 98.73% | 1.38s | 9.02s | 97 tps | 262K | $0.600 / $3.60 |
| Healthy | 100%◆ (best) | 99.85% | 98.47% | 99.11% | 801ms | 6.34s | 52 tps | 262K | $0.500 / $3.60 | |
Phalaphala | Healthy | 99.47% | 99.64% | 99.28% | 99.04% | 1.38s | 4.96s | 38 tps | 262K | $0.550 / $3.50 |
| Degraded | — | 95.62% | 97.66% | 97.47% | 1.43s | 10.2s | 94 tps | 262K | $0.550 / $3.50 | |
DigitalOceandigitalocean | No data | — | 94.69% | 93.59% | 93.76% | 1.03s | 8.80s | 6.0 tps | 131K | $0.550 / $3.50 |
| No data | — | 82.07% | 89.04% | 87.22% | 1.85s | 2.22s◆ (best) | 89 tps | 262K | $0.600 / $3.60 | |
| No data | — | — | — | — | — | — | — | 262K | $0.500 / $3.50 | |
StreamLakestreamlake | No data | — | 99.05% | 98.87% | 97.49% | 736ms◆ (best) | 3.36s | 92 tps | 256K | $0.600 / $3.60 |
Venicevenice | No data | — | 92.73% | 96.58% | 93.72% | 754ms | 4.14s | 54 tps | 128K | $0.750 / $4.50 |
- Alibaba100%◆ (best)Healthy24h 99.85%
- 7d
- 99.53%◆ (best)
- 30d
- 99.75%◆ (best)
- p50
- 962ms
- p99
- 5.65s
- Throughput
- 101 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.390 / $2.34◆ (best)
alibaba - DeepInfra100%◆ (best)Healthy24h 99.72%
- 7d
- 99.08%
- 30d
- 97.58%
- p50
- 993ms
- p99
- 2.28s
- Throughput
- 53 tps
- Context
- 262K
- $/Mtok
- $0.450 / $3.00
deepinfra/fp8fp8 - Novita100%◆ (best)Healthy24h 99.39%
- 7d
- 98.21%
- 30d
- 98.73%
- p50
- 1.38s
- p99
- 9.02s
- Throughput
- 97 tps
- Context
- 262K
- $/Mtok
- $0.600 / $3.60
novita - Parasail100%◆ (best)Healthy24h 99.85%
- 7d
- 98.47%
- 30d
- 99.11%
- p50
- 801ms
- p99
- 6.34s
- Throughput
- 52 tps
- Context
- 262K
- $/Mtok
- $0.500 / $3.60
parasail/fp8fp8 - Phala99.47%Healthy24h 99.64%
- 7d
- 99.28%
- 30d
- 99.04%
- p50
- 1.38s
- p99
- 4.96s
- Throughput
- 38 tps
- Context
- 262K
- $/Mtok
- $0.550 / $3.50
phala - Degraded24h 95.62%
- 7d
- 97.66%
- 30d
- 97.47%
- p50
- 1.43s
- p99
- 10.2s
- Throughput
- 94 tps
- Context
- 262K
- $/Mtok
- $0.550 / $3.50
atlas-cloud/fp8fp8 - No data24h 94.69%
- 7d
- 93.59%
- 30d
- 93.76%
- p50
- 1.03s
- p99
- 8.80s
- Throughput
- 6.0 tps
- Context
- 131K
- $/Mtok
- $0.550 / $3.50
digitalocean - No data24h 82.07%
- 7d
- 89.04%
- 30d
- 87.22%
- p50
- 1.85s
- p99
- 2.22s◆ (best)
- Throughput
- 89 tps
- Context
- 262K
- $/Mtok
- $0.600 / $3.60
gmicloud/fp8fp8 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.500 / $3.50
morph/fp4fp4 - No data24h 99.05%
- 7d
- 98.87%
- 30d
- 97.49%
- p50
- 736ms◆ (best)
- p99
- 3.36s
- Throughput
- 92 tps
- Context
- 256K
- $/Mtok
- $0.600 / $3.60
streamlake - No data24h 92.73%
- 7d
- 96.58%
- 30d
- 93.72%
- p50
- 754ms
- p99
- 4.14s
- Throughput
- 54 tps
- Context
- 128K
- $/Mtok
- $0.750 / $4.50
venice
Measured 1m 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.