Qwen3 Coder 480B A35B
qwen/qwen3-coder
Compare with Qwen3 235B A22B Instruct 2507, Qwen3 30B A3B Instruct 2507, Qwen3 Next 80B A3B Instruct
CompareQwen3 Coder 480B A35B
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba/opensource | Healthy | 100% | 99.91% | 99.87% | 99.87% | 952ms | 7.17s | 13 tps | 262K | $0.975 / $4.88 |
Googlegoogle-vertex/us-south1 | Healthy | 100% | 99.40% | 99.30% | 99.30% | 895ms | 4.33s | 69 tps | 262K | $0.220 / $1.80 |
| Down | 78.05% | 92.31% | 85.75% | 85.75% | 20.5s | 68.7s | 3.0 tps | 256K | $0.350 / $1.50 | |
| Down | — | 96.46% | 94.76% | 94.76% | 3.75s | 80.2s | 12 tps | 262K | $0.300 / $1.00 | |
| Degraded | — | 92.72% | 91.74% | 91.74% | 1.68s | 15.0s | 25 tps | 262K | $0.380 / $1.55 |
- Alibaba100%Healthy24h 99.91%
- 7d
- 99.87%
- 30d
- 99.87%
- p50
- 952ms
- p99
- 7.17s
- Throughput
- 13 tps
- Context
- 262K
- $/Mtok
- $0.975 / $4.88
alibaba/opensource - Google100%Healthy24h 99.40%
- 7d
- 99.30%
- 30d
- 99.30%
- p50
- 895ms
- p99
- 4.33s
- Throughput
- 69 tps
- Context
- 262K
- $/Mtok
- $0.220 / $1.80
google-vertex/us-south1 - Venice78.05%Down24h 92.31%
- 7d
- 85.75%
- 30d
- 85.75%
- p50
- 20.5s
- p99
- 68.7s
- Throughput
- 3.0 tps
- Context
- 256K
- $/Mtok
- $0.350 / $1.50
venice/fp8fp8 - Down24h 96.46%
- 7d
- 94.76%
- 30d
- 94.76%
- p50
- 3.75s
- p99
- 80.2s
- Throughput
- 12 tps
- Context
- 262K
- $/Mtok
- $0.300 / $1.00
deepinfra/turbofp4 - Degraded24h 92.72%
- 7d
- 91.74%
- 30d
- 91.74%
- p50
- 1.68s
- p99
- 15.0s
- Throughput
- 25 tps
- Context
- 262K
- $/Mtok
- $0.380 / $1.55
novita/fp8fp8
Measured 5m 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.
Throughput
Median tokens per second.