Qwen3 32B: 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) | 99.98% | 99.75% | 99.75% | 386ms | 2.93s◆ (best) | 25 tps | 41K | $0.080 / $0.280◆ (best) | |
Groqgroq | Healthy | 100%◆ (best) | 99.99% | 99.96%◆ (best) | 99.96%◆ (best) | 384ms◆ (best) | 5.35s | 227 tps◆ (best) | 131K | $0.290 / $0.590 |
| Degraded | 94.47% | 98.49% | 98.46% | 98.46% | 449ms | 4.66s | 21 tps | 41K | $0.100 / $0.300 | |
| Down | 36.26% | 91.03% | 88.82% | 88.82% | 213s | 300s | 4.0 tps | 131K | $0.140 / $0.570 |
- DeepInfra100%◆ (best)Healthy24h 99.98%
- 7d
- 99.75%
- 30d
- 99.75%
- p50
- 386ms
- p99
- 2.93s◆ (best)
- Throughput
- 25 tps
- Context
- 41K
- $/Mtok
- $0.080 / $0.280◆ (best)
deepinfra/fp8fp8 - Groq100%◆ (best)Healthy24h 99.99%
- 7d
- 99.96%◆ (best)
- 30d
- 99.96%◆ (best)
- p50
- 384ms◆ (best)
- p99
- 5.35s
- Throughput
- 227 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.290 / $0.590
groq - Nebius94.47%Degraded24h 98.49%
- 7d
- 98.46%
- 30d
- 98.46%
- p50
- 449ms
- p99
- 4.66s
- Throughput
- 21 tps
- Context
- 41K
- $/Mtok
- $0.100 / $0.300
nebius/basefp8 - SiliconFlow36.26%Down24h 91.03%
- 7d
- 88.82%
- 30d
- 88.82%
- p50
- 213s
- p99
- 300s
- Throughput
- 4.0 tps
- Context
- 131K
- $/Mtok
- $0.140 / $0.570
siliconflow/fp8fp8
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, averaged over the hour.
Throughput
Median tokens per second.