Qwen3 Coder 480B A35B
qwen/qwen3-coderInputOutput
Compare its 5 providersIncident RSSCompare with Qwen3 235B A22B Instruct 2507, Qwen3 30B A3B Instruct 2507, Qwen3 VL 235B A22B Instruct
CompareQwen3 Coder 480B A35B
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Googlegoogle-vertex/us-south1 | Healthy | 100% | 99.85% | 99.80% | 99.89% | 569ms | 3.07s | 71 tps | 262K | $0.220 / $1.80 |
Alibabaalibaba/opensource | No data | — | 99.99% | 99.96% | 99.75% | 1.25s | 1.89s | 38 tps | 262K | $0.975 / $4.88 |
| Degraded | — | 98.66% | 96.05% | 96.03% | 632ms | 19.8s | 63 tps | 262K | $0.300 / $1.00 | |
| No data | — | 93.47% | 63.71% | 84.46% | 936ms | 2.56s | 33 tps | 262K | $0.380 / $1.55 | |
| No data | — | 97.02% | 90.08% | 89.57% | 865ms | 8.28s | 58 tps | 256K | $0.350 / $1.50 |
- Google100%Healthy24h 99.85%
- 7d
- 99.80%
- 30d
- 99.89%
- p50
- 569ms
- p99
- 3.07s
- Throughput
- 71 tps
- Context
- 262K
- $/Mtok
- $0.220 / $1.80
google-vertex/us-south1 - No data24h 99.99%
- 7d
- 99.96%
- 30d
- 99.75%
- p50
- 1.25s
- p99
- 1.89s
- Throughput
- 38 tps
- Context
- 262K
- $/Mtok
- $0.975 / $4.88
alibaba/opensource - Degraded24h 98.66%
- 7d
- 96.05%
- 30d
- 96.03%
- p50
- 632ms
- p99
- 19.8s
- Throughput
- 63 tps
- Context
- 262K
- $/Mtok
- $0.300 / $1.00
deepinfra/turbofp4 - No data24h 93.47%
- 7d
- 63.71%
- 30d
- 84.46%
- p50
- 936ms
- p99
- 2.56s
- Throughput
- 33 tps
- Context
- 262K
- $/Mtok
- $0.380 / $1.55
novita/fp8fp8 - No data24h 97.02%
- 7d
- 90.08%
- 30d
- 89.57%
- p50
- 865ms
- p99
- 8.28s
- Throughput
- 58 tps
- Context
- 256K
- $/Mtok
- $0.350 / $1.50
venice/fp8fp8
Measured 1m 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 · 5 endpoints · 47 days with data, 128 endpoint-days below 90%.