Qwen3.8 2.4T A95B
qwen/qwen3.8-2.4t-a95b
Compare with Qwen3.8 27B, Qwen-Plus, Qwen3.8 Max
CompareQwen3.8 2.4T A95B
| Provider | State | Uptime 5m | 24h | p50 latency | p99 | Throughput | Context | $/Mtok |
|---|---|---|---|---|---|---|---|---|
Modalmodal/nvfp4nvfp4 | Healthy | 100% | 99.30% | 1.55s | 7.79s | 73 tps | 1.0M | $2.00 / $6.00 |
DeepInfradeepinfra/fp4fp4 | Healthy | 98.28% | 95.19% | 1.12s | 9.01s | 56 tps | 262K | $2.00 / $6.00 |
Alibabaalibaba | No data | — | 99.95% | 3.07s | 14.4s | 34 tps | 1.0M | $2.00 / $6.00 |
DigitalOceandigitalocean | No data | — | 95.21% | — | — | — | 262K | $2.00 / $6.00 |
SiliconFlowsiliconflow/fp8fp8 | No data | — | 99.26% | 2.79s | 15.7s | 44 tps | 1.0M | $2.00 / $6.00 |
Togethertogether | No data | — | 99.84% | 4.71s | 30.1s | 85 tps | 1.0M | $2.50 / $6.25 |
Venicevenice | No data | — | 98.23% | 1.85s | 17.0s | 77 tps | 262K | $2.50 / $7.50 |
- Modal100%Healthy24h 99.30%
- p50
- 1.55s
- p99
- 7.79s
- Throughput
- 73 tps
- Context
- 1.0M
- $/Mtok
- $2.00 / $6.00
modal/nvfp4nvfp4 - DeepInfra98.28%Healthy24h 95.19%
- p50
- 1.12s
- p99
- 9.01s
- Throughput
- 56 tps
- Context
- 262K
- $/Mtok
- $2.00 / $6.00
deepinfra/fp4fp4 - Alibaba—No data24h 99.95%
- p50
- 3.07s
- p99
- 14.4s
- Throughput
- 34 tps
- Context
- 1.0M
- $/Mtok
- $2.00 / $6.00
alibaba - DigitalOcean—No data24h 95.21%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 262K
- $/Mtok
- $2.00 / $6.00
digitalocean - SiliconFlow—No data24h 99.26%
- p50
- 2.79s
- p99
- 15.7s
- Throughput
- 44 tps
- Context
- 1.0M
- $/Mtok
- $2.00 / $6.00
siliconflow/fp8fp8 - Together—No data24h 99.84%
- p50
- 4.71s
- p99
- 30.1s
- Throughput
- 85 tps
- Context
- 1.0M
- $/Mtok
- $2.50 / $6.25
together - Venice—No data24h 98.23%
- p50
- 1.85s
- p99
- 17.0s
- Throughput
- 77 tps
- Context
- 262K
- $/Mtok
- $2.50 / $7.50
venice
Measured 4m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
5 endpoints reporting no data are omitted from the charts.
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 and p99 are OpenRouter's rolling 30-minute windows.
Throughput
Median tokens per second.