Qwen3 14B: 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 | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
Alibabaalibaba | Healthy | 100%◆ (best) | 99.85% | 99.96%◆ (best) | 99.96%◆ (best) | 215ms◆ (best) | 751ms◆ (best) | 62 tps◆ (best) | 131K | $0.227 / $0.910 |
| Healthy | 100%◆ (best) | 98.95% | 97.05% | 97.05% | 700ms | 5.56s | 25 tps | 41K | $0.120 / $0.240 | |
| Healthy | 99.68% | 99.88% | 99.90% | 99.90% | 852ms | 7.74s | 37 tps | 41K | $0.100 / $0.220◆ (best) |
- Alibaba100%◆ (best)Healthy24h 99.85%
- 7d
- 99.96%◆ (best)
- 30d
- 99.96%◆ (best)
- p50
- 215ms◆ (best)
- p99
- 751ms◆ (best)
- Throughput
- 62 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.227 / $0.910
alibaba - DeepInfra100%◆ (best)Healthy24h 98.95%
- 7d
- 97.05%
- 30d
- 97.05%
- p50
- 700ms
- p99
- 5.56s
- Throughput
- 25 tps
- Context
- 41K
- $/Mtok
- $0.120 / $0.240
deepinfra/fp8fp8 - NextBit99.68%Healthy24h 99.88%
- 7d
- 99.90%
- 30d
- 99.90%
- p50
- 852ms
- p99
- 7.74s
- Throughput
- 37 tps
- Context
- 41K
- $/Mtok
- $0.100 / $0.220◆ (best)
nextbit/int4int4
Measured 5m 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.