Qwen3 VL 235B A22B 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) | 97.36% | 95.86% | 95.75% | 1.62s | 40.9s | 35 tps | 131K | $0.300 / $1.50 | |
| Healthy | 100%◆ (best) | 99.31% | 99.10%◆ (best) | 99.11% | 889ms | 6.24s | 36 tps | 131K | $0.210 / $1.90 | |
| Healthy | 97.50% | 98.40% | 96.62% | 95.87% | 2.99s | 25.9s | 10 tps | 262K | $0.200 / $0.880◆ (best) | |
Alibabaalibaba | Healthy | 96.03% | 99.20% | 98.98% | 99.45%◆ (best) | 612ms◆ (best) | 2.13s◆ (best) | 43 tps◆ (best) | 131K | $0.260 / $1.04 |
| Healthy | 95.73% | 99.28% | 96.58% | 96.36% | 1.62s | 14.1s | 42 tps | 128K | $0.210 / $1.90 | |
PrimeIntellectprimeintellect | No data | — | — | 98.03% | 98.03% | — | — | — | 262K | $0.300 / $1.50 |
- Novita100%◆ (best)Healthy24h 97.36%
- 7d
- 95.86%
- 30d
- 95.75%
- p50
- 1.62s
- p99
- 40.9s
- Throughput
- 35 tps
- Context
- 131K
- $/Mtok
- $0.300 / $1.50
novita/bf16bf16 - Parasail100%◆ (best)Healthy24h 99.31%
- 7d
- 99.10%◆ (best)
- 30d
- 99.11%
- p50
- 889ms
- p99
- 6.24s
- Throughput
- 36 tps
- Context
- 131K
- $/Mtok
- $0.210 / $1.90
parasail/fp8fp8 - DeepInfra97.50%Healthy24h 98.40%
- 7d
- 96.62%
- 30d
- 95.87%
- p50
- 2.99s
- p99
- 25.9s
- Throughput
- 10 tps
- Context
- 262K
- $/Mtok
- $0.200 / $0.880◆ (best)
deepinfra/fp8fp8 - Alibaba96.03%Healthy24h 99.20%
- 7d
- 98.98%
- 30d
- 99.45%◆ (best)
- p50
- 612ms◆ (best)
- p99
- 2.13s◆ (best)
- Throughput
- 43 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.260 / $1.04
alibaba - Venice95.73%Healthy24h 99.28%
- 7d
- 96.58%
- 30d
- 96.36%
- p50
- 1.62s
- p99
- 14.1s
- Throughput
- 42 tps
- Context
- 128K
- $/Mtok
- $0.210 / $1.90
venice/fp8fp8 - No data24h —
- 7d
- 98.03%
- 30d
- 98.03%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.300 / $1.50
primeintellect
Measured 4m 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.