Qwen3 VL 30B A3B Instruct: 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) | 99.74% | 99.18% | 99.18% | 416ms◆ (best) | 3.35s◆ (best) | 27 tps | 262K | $0.150 / $0.600 | |
Alibabaalibaba | Healthy | 99.86% | 99.92% | 99.91%◆ (best) | 99.91%◆ (best) | 546ms | 4.14s | 36 tps◆ (best) | 131K | $0.130 / $0.520◆ (best) |
| Degraded | 96.67% | 98.10% | 97.80% | 97.80% | 2.47s | 6.14s | 13 tps | 262K | $0.290 / $1.00 | |
| Degraded | 93.75% | 96.19% | 96.19% | 96.19% | 1.01s | 8.18s | 26 tps | 131K | $0.200 / $0.700 |
- DeepInfra100%◆ (best)Healthy24h 99.74%
- 7d
- 99.18%
- 30d
- 99.18%
- p50
- 416ms◆ (best)
- p99
- 3.35s◆ (best)
- Throughput
- 27 tps
- Context
- 262K
- $/Mtok
- $0.150 / $0.600
deepinfra/fp8fp8 - Alibaba99.86%Healthy24h 99.92%
- 7d
- 99.91%◆ (best)
- 30d
- 99.91%◆ (best)
- p50
- 546ms
- p99
- 4.14s
- Throughput
- 36 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.130 / $0.520◆ (best)
alibaba - SiliconFlow96.67%Degraded24h 98.10%
- 7d
- 97.80%
- 30d
- 97.80%
- p50
- 2.47s
- p99
- 6.14s
- Throughput
- 13 tps
- Context
- 262K
- $/Mtok
- $0.290 / $1.00
siliconflow/fp8fp8 - Novita93.75%Degraded24h 96.19%
- 7d
- 96.19%
- 30d
- 96.19%
- p50
- 1.01s
- p99
- 8.18s
- Throughput
- 26 tps
- Context
- 131K
- $/Mtok
- $0.200 / $0.700
novita/bf16bf16
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, averaged over the hour.
Throughput
Median tokens per second.