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 | 99.86%◆ (best) | 99.65% | 99.75%◆ (best) | 91.35% | 1.62s | 6.65s | 18 tps | 262K | $0.150 / $0.600 | |
Alibabaalibaba | Healthy | 99.13% | 99.33% | 99.39% | 99.75%◆ (best) | 597ms◆ (best) | 2.68s◆ (best) | 59 tps | 131K | $0.130 / $0.520 |
| Healthy | 97.00% | 97.73% | 98.69% | 97.74% | 1.11s | 6.31s | 66 tps◆ (best) | 131K | $0.200 / $0.700 | |
| Down | 95.52% | 87.66% | 69.51% | 83.29% | 1.29s | 5.40s | 17 tps | 262K | $0.290 / $1.00 | |
| No data | — | — | — | — | — | — | — | 131K | $0.090 / $0.400◆ (best) | |
PrimeIntellectprimeintellect | No data | — | — | 99.09% | 99.09% | — | — | — | 262K | $0.200 / $0.700 |
- DeepInfra99.86%◆ (best)Healthy24h 99.65%
- 7d
- 99.75%◆ (best)
- 30d
- 91.35%
- p50
- 1.62s
- p99
- 6.65s
- Throughput
- 18 tps
- Context
- 262K
- $/Mtok
- $0.150 / $0.600
deepinfra/fp8fp8 - Alibaba99.13%Healthy24h 99.33%
- 7d
- 99.39%
- 30d
- 99.75%◆ (best)
- p50
- 597ms◆ (best)
- p99
- 2.68s◆ (best)
- Throughput
- 59 tps
- Context
- 131K
- $/Mtok
- $0.130 / $0.520
alibaba - Novita97.00%Healthy24h 97.73%
- 7d
- 98.69%
- 30d
- 97.74%
- p50
- 1.11s
- p99
- 6.31s
- Throughput
- 66 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.200 / $0.700
novita/bf16bf16 - SiliconFlow95.52%Down24h 87.66%
- 7d
- 69.51%
- 30d
- 83.29%
- p50
- 1.29s
- p99
- 5.40s
- Throughput
- 17 tps
- Context
- 262K
- $/Mtok
- $0.290 / $1.00
siliconflow/fp8fp8 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 131K
- $/Mtok
- $0.090 / $0.400◆ (best)
darkbloom/fp4fp4 - No data24h —
- 7d
- 99.09%
- 30d
- 99.09%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.200 / $0.700
primeintellect
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.