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 | No data | — | — | 99.53%◆ (best) | 99.75%◆ (best) | — | — | — | 262K | $0.390 / $2.34◆ (best) |
| No data | — | — | 97.64% | 97.46% | — | — | — | 262K | $0.550 / $3.50 | |
| No data | — | — | 99.09% | 97.58% | — | — | — | 262K | $0.450 / $3.00 | |
DigitalOceandigitalocean | No data | — | — | 93.61% | 93.76% | — | — | — | 131K | $0.550 / $3.50 |
| No data | — | — | 88.53% | 87.11% | — | — | — | 262K | $0.600 / $3.60 | |
| No data | — | — | — | — | — | — | — | 262K | $0.500 / $3.50 | |
Novitanovita | No data | — | — | 98.22% | 98.74% | — | — | — | 262K | $0.600 / $3.60 |
| No data | — | — | 98.47% | 99.11% | — | — | — | 262K | $0.500 / $3.60 | |
Phalaphala | No data | — | — | 99.29% | 99.04% | — | — | — | 262K | $0.550 / $3.50 |
StreamLakestreamlake | No data | — | — | 98.87% | 97.50% | — | — | — | 256K | $0.600 / $3.60 |
Venicevenice | No data | — | — | 96.58% | 93.72% | — | — | — | 128K | $0.750 / $4.50 |
- No data24h —
- 7d
- 99.53%◆ (best)
- 30d
- 99.75%◆ (best)
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.390 / $2.34◆ (best)
alibaba - No data24h —
- 7d
- 97.64%
- 30d
- 97.46%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.550 / $3.50
atlas-cloud/fp8fp8 - No data24h —
- 7d
- 99.09%
- 30d
- 97.58%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.450 / $3.00
deepinfra/fp8fp8 - No data24h —
- 7d
- 93.61%
- 30d
- 93.76%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.550 / $3.50
digitalocean - No data24h —
- 7d
- 88.53%
- 30d
- 87.11%
- p50
- —
- p99
- —
- Throughput
- —
- 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 —
- 7d
- 98.22%
- 30d
- 98.74%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.600 / $3.60
novita - No data24h —
- 7d
- 98.47%
- 30d
- 99.11%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.500 / $3.60
parasail/fp8fp8 - No data24h —
- 7d
- 99.29%
- 30d
- 99.04%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.550 / $3.50
phala - No data24h —
- 7d
- 98.87%
- 30d
- 97.50%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 256K
- $/Mtok
- $0.600 / $3.60
streamlake - No data24h —
- 7d
- 96.58%
- 30d
- 93.72%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 128K
- $/Mtok
- $0.750 / $4.50
venice
No recent measurement. ◆ marks the best value in each column among these endpoints.
Price vs latency
No endpoint currently reports both a price and a latency.
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.