Qwen3.5-35B-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) | 99.94% | 99.89% | 99.89% | 1.04s | 2.97s | 80 tps | 262K | $0.225 / $1.80 | |
| Healthy | 100%◆ (best) | 99.99% | 99.96%◆ (best) | 99.96%◆ (best) | 354ms◆ (best) | 8.80s | 128 tps◆ (best) | 262K | $0.250 / $1.25 | |
| Healthy | 100%◆ (best) | 99.92% | 99.74% | 99.74% | 604ms | 22.3s | 87 tps | 262K | $0.150 / $1.00 | |
| Healthy | 100%◆ (best) | 99.26% | 97.58% | 97.58% | 1.48s | 16.5s | 32 tps | 262K | $0.240 / $1.80 | |
Venicevenice | Healthy | 100%◆ (best) | 99.92% | 99.90% | 99.90% | 907ms | 3.91s | 119 tps | 256K | $0.313 / $1.25 |
| Healthy | 95.00% | 99.80% | 98.13% | 98.13% | 438ms | 1.87s◆ (best) | 60 tps | 262K | $0.140 / $1.00◆ (best) | |
Alibabaalibaba | Degraded | 94.59% | 99.20% | 99.29% | 99.29% | 614ms | 6.21s | 102 tps | 262K | $0.163 / $1.30 |
| No data | — | — | 99.60% | 99.60% | — | — | — | 262K | $0.140 / $1.00◆ (best) |
- AtlasCloud100%◆ (best)Healthy24h 99.94%
- 7d
- 99.89%
- 30d
- 99.89%
- p50
- 1.04s
- p99
- 2.97s
- Throughput
- 80 tps
- Context
- 262K
- $/Mtok
- $0.225 / $1.80
atlas-cloud/fp8fp8 - CoreWeave100%◆ (best)Healthy24h 99.99%
- 7d
- 99.96%◆ (best)
- 30d
- 99.96%◆ (best)
- p50
- 354ms◆ (best)
- p99
- 8.80s
- Throughput
- 128 tps◆ (best)
- Context
- 262K
- $/Mtok
- $0.250 / $1.25
coreweave/fp8fp8 - Parasail100%◆ (best)Healthy24h 99.92%
- 7d
- 99.74%
- 30d
- 99.74%
- p50
- 604ms
- p99
- 22.3s
- Throughput
- 87 tps
- Context
- 262K
- $/Mtok
- $0.150 / $1.00
parasail/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 99.26%
- 7d
- 97.58%
- 30d
- 97.58%
- p50
- 1.48s
- p99
- 16.5s
- Throughput
- 32 tps
- Context
- 262K
- $/Mtok
- $0.240 / $1.80
siliconflow/fp8fp8 - Venice100%◆ (best)Healthy24h 99.92%
- 7d
- 99.90%
- 30d
- 99.90%
- p50
- 907ms
- p99
- 3.91s
- Throughput
- 119 tps
- Context
- 256K
- $/Mtok
- $0.313 / $1.25
venice - DeepInfra95.00%Healthy24h 99.80%
- 7d
- 98.13%
- 30d
- 98.13%
- p50
- 438ms
- p99
- 1.87s◆ (best)
- Throughput
- 60 tps
- Context
- 262K
- $/Mtok
- $0.140 / $1.00◆ (best)
deepinfra/fp8fp8 - Alibaba94.59%Degraded24h 99.20%
- 7d
- 99.29%
- 30d
- 99.29%
- p50
- 614ms
- p99
- 6.21s
- Throughput
- 102 tps
- Context
- 262K
- $/Mtok
- $0.163 / $1.30
alibaba - No data24h —
- 7d
- 99.60%
- 30d
- 99.60%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $0.140 / $1.00◆ (best)
akashml/fp8fp8
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, averaged over the hour.
Throughput
Median tokens per second.