Qwen3 30B A3B Instruct 2507: 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) | 99.75% | 99.60%◆ (best) | 99.81%◆ (best) | 347ms | 968ms◆ (best) | 66 tps | 131K | $0.130 / $0.520 |
StreamLakestreamlake | Healthy | 99.59% | 99.62% | 99.01% | 98.90% | 727ms | 4.51s | 85 tps◆ (best) | 128K | $0.048 / $0.193◆ (best) |
DekaLLMdekallm | Healthy | 98.60% | 98.89% | 98.71% | 98.93% | 456ms | 3.42s | 79 tps | 262K | $0.090 / $0.300 |
| Healthy | 98.40% | 98.49% | 98.83% | 97.80% | 1.05s | 5.87s | 19 tps | 262K | $0.090 / $0.300 | |
| Healthy | 97.99% | 91.25% | 94.05% | 93.83% | 332ms◆ (best) | 4.69s | 31 tps | 262K | $0.100 / $0.300 | |
| No data | — | — | — | — | — | — | — | 262K | $0.100 / $0.300 | |
PrimeIntellectprimeintellect | No data | — | — | 95.97% | 95.97% | — | — | — | 262K | $0.130 / $0.520 |
- Alibaba100%◆ (best)Healthy24h 99.75%
- 7d
- 99.60%◆ (best)
- 30d
- 99.81%◆ (best)
- p50
- 347ms
- p99
- 968ms◆ (best)
- Throughput
- 66 tps
- Context
- 131K
- $/Mtok
- $0.130 / $0.520
alibaba - StreamLake99.59%Healthy24h 99.62%
- 7d
- 99.01%
- 30d
- 98.90%
- p50
- 727ms
- p99
- 4.51s
- Throughput
- 85 tps◆ (best)
- Context
- 128K
- $/Mtok
- $0.048 / $0.193◆ (best)
streamlake - DekaLLM98.60%Healthy24h 98.89%
- 7d
- 98.71%
- 30d
- 98.93%
- p50
- 456ms
- p99
- 3.42s
- Throughput
- 79 tps
- Context
- 262K
- $/Mtok
- $0.090 / $0.300
dekallm - SiliconFlow98.40%Healthy24h 98.49%
- 7d
- 98.83%
- 30d
- 97.80%
- p50
- 1.05s
- p99
- 5.87s
- Throughput
- 19 tps
- Context
- 262K
- $/Mtok
- $0.090 / $0.300
siliconflow/fp8fp8 - Nebius97.99%Healthy24h 91.25%
- 7d
- 94.05%
- 30d
- 93.83%
- p50
- 332ms◆ (best)
- p99
- 4.69s
- Throughput
- 31 tps
- Context
- 262K
- $/Mtok
- $0.100 / $0.300
nebius/fp8fp8 - No data24h —
- 7d
- —
- 30d
- —
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.100 / $0.300
coreweave/bf16bf16 - No data24h —
- 7d
- 95.97%
- 30d
- 95.97%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.130 / $0.520
primeintellect
Measured 1m 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.