Qwen3.6 35B A3B: 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) | 99.40% | 99.69% | 99.69% | 598ms | 3.67s◆ (best) | 83 tps | 262K | $0.140 / $1.00 | |
| Healthy | 100%◆ (best) | 98.86% | 99.30% | 99.30% | 875ms | 9.64s | 103 tps | 262K | $0.186 / $1.11 | |
| Healthy | 100%◆ (best) | 99.33% | 99.65% | 99.65% | 449ms◆ (best) | 5.31s | 136 tps◆ (best) | 262K | $0.250 / $1.25 | |
| Healthy | 100%◆ (best) | 97.73% | 93.85% | 93.85% | 754ms | 21.4s | 54 tps | 262K | $0.100 / $0.950 | |
| Healthy | 100%◆ (best) | 99.42% | 99.38% | 99.38% | 541ms | 9.01s | 133 tps | 262K | $0.190 / $1.19 | |
| Healthy | 100%◆ (best) | 98.78% | 99.20% | 99.20% | 538ms | 4.73s | 53 tps | 262K | $0.150 / $1.00 | |
| Healthy | 100%◆ (best) | 97.06% | 98.01% | 98.01% | 1.97s | 23.7s | 35 tps | 262K | $0.200 / $1.60 | |
| Healthy | 100%◆ (best) | 99.68% | 99.85%◆ (best) | 99.85%◆ (best) | 661ms | 4.77s | 103 tps | 256K | $0.100 / $1.00 | |
Phalaphala | Degraded | 96.19% | 96.50% | 98.16% | 98.16% | 1.50s | 9.43s | 110 tps | 262K | $0.200 / $1.27 |
| Down | 87.61% | 86.87% | 87.93% | 87.93% | 4.57s | 23.4s | 40 tps | 262K | $0.080 / $0.750◆ (best) | |
Darkbloomdarkbloom | No data | — | — | 69.81% | 69.81% | — | — | — | 262K | $0.080 / $0.750◆ (best) |
- AkashML100%◆ (best)Healthy24h 99.40%
- 7d
- 99.69%
- 30d
- 99.69%
- p50
- 598ms
- p99
- 3.67s◆ (best)
- Throughput
- 83 tps
- Context
- 262K
- $/Mtok
- $0.140 / $1.00
akashml/fp8fp8 - AtlasCloud100%◆ (best)Healthy24h 98.86%
- 7d
- 99.30%
- 30d
- 99.30%
- p50
- 875ms
- p99
- 9.64s
- Throughput
- 103 tps
- Context
- 262K
- $/Mtok
- $0.186 / $1.11
atlas-cloud/fp8fp8 - CoreWeave100%◆ (best)Healthy24h 99.33%
- 7d
- 99.65%
- 30d
- 99.65%
- p50
- 449ms◆ (best)
- p99
- 5.31s
- Throughput
- 136 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.250 / $1.25
coreweave/fp8fp8 - DeepInfra100%◆ (best)Healthy24h 97.73%
- 7d
- 93.85%
- 30d
- 93.85%
- p50
- 754ms
- p99
- 21.4s
- Throughput
- 54 tps
- Context
- 262K
- $/Mtok
- $0.100 / $0.950
deepinfra/fp8fp8 - Io Net100%◆ (best)Healthy24h 99.42%
- 7d
- 99.38%
- 30d
- 99.38%
- p50
- 541ms
- p99
- 9.01s
- Throughput
- 133 tps
- Context
- 262K
- $/Mtok
- $0.190 / $1.19
io-net/fp8fp8 - Parasail100%◆ (best)Healthy24h 98.78%
- 7d
- 99.20%
- 30d
- 99.20%
- p50
- 538ms
- p99
- 4.73s
- Throughput
- 53 tps
- Context
- 262K
- $/Mtok
- $0.150 / $1.00
parasail/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 97.06%
- 7d
- 98.01%
- 30d
- 98.01%
- p50
- 1.97s
- p99
- 23.7s
- Throughput
- 35 tps
- Context
- 262K
- $/Mtok
- $0.200 / $1.60
siliconflow/fp8fp8 - Venice100%◆ (best)Healthy24h 99.68%
- 7d
- 99.85%◆ (best)
- 30d
- 99.85%◆ (best)
- p50
- 661ms
- p99
- 4.77s
- Throughput
- 103 tps
- Context
- 256K
- $/Mtok
- $0.100 / $1.00
venice/fp8fp8 - Phala96.19%Degraded24h 96.50%
- 7d
- 98.16%
- 30d
- 98.16%
- p50
- 1.50s
- p99
- 9.43s
- Throughput
- 110 tps
- Context
- 262K
- $/Mtok
- $0.200 / $1.27
phala - Darkbloom87.61%Down24h 86.87%
- 7d
- 87.93%
- 30d
- 87.93%
- p50
- 4.57s
- p99
- 23.4s
- Throughput
- 40 tps
- Context
- 262K
- $/Mtok
- $0.080 / $0.750◆ (best)
darkbloom/fp4fp4 - No data24h —
- 7d
- 69.81%
- 30d
- 69.81%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.080 / $0.750◆ (best)
darkbloom
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.
Throughput
Median tokens per second.