Qwen3 VL 235B A22B Instruct
qwen/qwen3-vl-235b-a22b-instructInputOutput
Compare its 6 providersIncident RSSCompare with Qwen3 235B A22B Instruct 2507, Qwen3 30B A3B Instruct 2507, Qwen3 Coder 480B A35B
CompareQwen3 VL 235B A22B Instruct
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100% | 99.25% | 98.95% | 99.46% | 655ms | 1.95s | 42 tps | 131K | $0.260 / $1.04 |
| Healthy | 100% | 99.41% | 99.14% | 99.10% | 1.14s | 7.31s | 38 tps | 131K | $0.210 / $1.90 | |
| Healthy | 99.50% | 97.01% | 96.85% | 95.82% | 2.01s | 14.6s | 18 tps | 262K | $0.200 / $0.880 | |
| No data | — | 96.30% | 96.05% | 95.79% | 1.87s | 8.76s | 27 tps | 131K | $0.300 / $1.50 | |
PrimeIntellectprimeintellect | No data | — | 98.83% | 92.89% | 92.89% | 1.39s | 3.33s | 34 tps | 262K | $0.300 / $1.50 |
| No data | — | 97.27% | 96.50% | 96.18% | 4.58s | 12.3s | 14 tps | 128K | $0.210 / $1.90 |
- Alibaba100%Healthy24h 99.25%
- 7d
- 98.95%
- 30d
- 99.46%
- p50
- 655ms
- p99
- 1.95s
- Throughput
- 42 tps
- Context
- 131K
- $/Mtok
- $0.260 / $1.04
alibaba - Parasail100%Healthy24h 99.41%
- 7d
- 99.14%
- 30d
- 99.10%
- p50
- 1.14s
- p99
- 7.31s
- Throughput
- 38 tps
- Context
- 131K
- $/Mtok
- $0.210 / $1.90
parasail/fp8fp8 - DeepInfra99.50%Healthy24h 97.01%
- 7d
- 96.85%
- 30d
- 95.82%
- p50
- 2.01s
- p99
- 14.6s
- Throughput
- 18 tps
- Context
- 262K
- $/Mtok
- $0.200 / $0.880
deepinfra/fp8fp8 - No data24h 96.30%
- 7d
- 96.05%
- 30d
- 95.79%
- p50
- 1.87s
- p99
- 8.76s
- Throughput
- 27 tps
- Context
- 131K
- $/Mtok
- $0.300 / $1.50
novita/bf16bf16 - No data24h 98.83%
- 7d
- 92.89%
- 30d
- 92.89%
- p50
- 1.39s
- p99
- 3.33s
- Throughput
- 34 tps
- Context
- 262K
- $/Mtok
- $0.300 / $1.50
primeintellect - No data24h 97.27%
- 7d
- 96.50%
- 30d
- 96.18%
- p50
- 4.58s
- p99
- 12.3s
- Throughput
- 14 tps
- Context
- 128K
- $/Mtok
- $0.210 / $1.90
venice/fp8fp8
Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
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.
Daily availability
One row per endpoint · one dot per day · last 90 days.
worst hour ≥ 95%90–95%< 90%no data
Colour by each endpoint's worst hour of each UTC day · 6 endpoints · 47 days with data, 81 endpoint-days below 90%.