DeepSeek V3.2
deepseek/deepseek-v3.2
Compare with DeepSeek V3, DeepSeek V3.2 Exp, DeepSeek V3.1
CompareDeepSeek V3.2
| Provider | State | Uptime 5m | 24h | p50 latency | p99 | Throughput | Context | $/Mtok |
|---|---|---|---|---|---|---|---|---|
Baidubaidu/fp8fp8 | Healthy | 100% | 99.92% | 1.08s | 4.17s | 42 tps | 131K | $0.280 / $0.420 |
DigitalOceandigitalocean | Degraded | 100% | 95.56% | 990ms | 19.2s | 22 tps | 164K | $0.250 / $0.800 |
Friendlifriendli | Healthy | 100% | 95.72% | 420ms | 6.44s | 59 tps | 164K | $0.500 / $1.50 |
Googlegoogle-vertex | Healthy | 100% | 99.70% | 965ms | 6.61s | 12 tps | 164K | $0.560 / $1.68 |
Novitanovita/fp8fp8 | Healthy | 100% | 99.99% | 1.40s | 5.17s | 27 tps | 164K | $0.269 / $0.400 |
Phalaphala | Healthy | 100% | 99.51% | 4.67s | 12.8s | 6.0 tps | 164K | $1.00 / $1.00 |
AtlasCloudatlas-cloud/fp8fp8 | Healthy | 99.98% | 99.92% | 1.47s | 7.40s | 19 tps | 164K | $0.260 / $0.380 |
StreamLakestreamlake/fp8fp8 | Healthy | 99.91% | 99.87% | 1.42s | 20.2s | 25 tps | 128K | $0.214 / $0.322 |
SiliconFlowsiliconflow/fp8fp8 | Healthy | 99.65% | 99.78% | 1.99s | 7.10s | 24 tps | 164K | $0.259 / $0.420 |
Alibabaalibaba/fp8fp8 | Healthy | 98.98% | 98.24% | 816ms | 10.3s | 53 tps | 131K | $0.370 / $1.11 |
GMICloudgmicloud/fp8fp8 | Healthy | 95.06% | 98.35% | 2.25s | 12.0s | 27 tps | 164K | $0.209 / $0.310 |
DeepInfradeepinfra/fp4fp4 | Degraded | 92.65% | 96.20% | 1.51s | 25.9s | 8.0 tps | 164K | $0.260 / $0.380 |
SambaNovasambanova | No data | — | 84.98% | — | — | — | 33K | $3.00 / $4.50 |
Venicevenice | No data | — | 92.00% | 1.70s | 20.7s | 10 tps | 160K | $0.330 / $0.480 |
- Baidu100%Healthy24h 99.92%
- p50
- 1.08s
- p99
- 4.17s
- Throughput
- 42 tps
- Context
- 131K
- $/Mtok
- $0.280 / $0.420
baidu/fp8fp8 - DigitalOcean100%Degraded24h 95.56%
- p50
- 990ms
- p99
- 19.2s
- Throughput
- 22 tps
- Context
- 164K
- $/Mtok
- $0.250 / $0.800
digitalocean - Friendli100%Healthy24h 95.72%
- p50
- 420ms
- p99
- 6.44s
- Throughput
- 59 tps
- Context
- 164K
- $/Mtok
- $0.500 / $1.50
friendli - Google100%Healthy24h 99.70%
- p50
- 965ms
- p99
- 6.61s
- Throughput
- 12 tps
- Context
- 164K
- $/Mtok
- $0.560 / $1.68
google-vertex - Novita100%Healthy24h 99.99%
- p50
- 1.40s
- p99
- 5.17s
- Throughput
- 27 tps
- Context
- 164K
- $/Mtok
- $0.269 / $0.400
novita/fp8fp8 - Phala100%Healthy24h 99.51%
- p50
- 4.67s
- p99
- 12.8s
- Throughput
- 6.0 tps
- Context
- 164K
- $/Mtok
- $1.00 / $1.00
phala - AtlasCloud99.98%Healthy24h 99.92%
- p50
- 1.47s
- p99
- 7.40s
- Throughput
- 19 tps
- Context
- 164K
- $/Mtok
- $0.260 / $0.380
atlas-cloud/fp8fp8 - StreamLake99.91%Healthy24h 99.87%
- p50
- 1.42s
- p99
- 20.2s
- Throughput
- 25 tps
- Context
- 128K
- $/Mtok
- $0.214 / $0.322
streamlake/fp8fp8 - SiliconFlow99.65%Healthy24h 99.78%
- p50
- 1.99s
- p99
- 7.10s
- Throughput
- 24 tps
- Context
- 164K
- $/Mtok
- $0.259 / $0.420
siliconflow/fp8fp8 - Alibaba98.98%Healthy24h 98.24%
- p50
- 816ms
- p99
- 10.3s
- Throughput
- 53 tps
- Context
- 131K
- $/Mtok
- $0.370 / $1.11
alibaba/fp8fp8 - GMICloud95.06%Healthy24h 98.35%
- p50
- 2.25s
- p99
- 12.0s
- Throughput
- 27 tps
- Context
- 164K
- $/Mtok
- $0.209 / $0.310
gmicloud/fp8fp8 - DeepInfra92.65%Degraded24h 96.20%
- p50
- 1.51s
- p99
- 25.9s
- Throughput
- 8.0 tps
- Context
- 164K
- $/Mtok
- $0.260 / $0.380
deepinfra/fp4fp4 - SambaNova—No data24h 84.98%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 33K
- $/Mtok
- $3.00 / $4.50
sambanova - Venice—No data24h 92.00%
- p50
- 1.70s
- p99
- 20.7s
- Throughput
- 10 tps
- Context
- 160K
- $/Mtok
- $0.330 / $0.480
venice
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
2 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, averaged over the hour.
Throughput
Median tokens per second.