DeepSeek V3.2
deepseek/deepseek-v3.2InputOutput
Compare its 15 providersIncident RSSCompare with DeepSeek V3, DeepSeek V3.2 Exp, DeepSeek V3.1
CompareDeepSeek V3.2
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 98.74% | 96.02% | 98.94% | 1.49s | 5.18s | 43 tps | 131K | $0.280 / $0.420 | |
| Healthy | 100% | 99.74% | 99.09% | 99.05% | 1.51s | 10.6s | 15 tps | 164K | $0.260 / $0.380 | |
Friendlifriendli | Healthy | 100% | 99.97% | 99.89% | 99.96% | 1.03s | 8.25s | 43 tps | 164K | $0.500 / $1.50 |
Googlegoogle-vertex | Healthy | 100% | 99.12% | 99.52% | 99.18% | 1.70s | 23.0s | 37 tps | 164K | $0.560 / $1.68 |
Phalaphala | Healthy | 100% | 99.57% | 99.56% | 99.42% | 2.97s | 13.1s | 11 tps | 164K | $1.00 / $1.00 |
Venicevenice | Healthy | 100% | 99.79% | 98.83% | 98.83% | 1.70s | 7.71s | 14 tps | 160K | $0.268 / $0.390 |
DigitalOceandigitalocean | Healthy | 99.97% | 98.04% | 97.56% | 97.81% | 915ms | 7.50s | 32 tps | 164K | $0.300 / $0.960 |
| Healthy | 97.15% | 95.73% | 96.20% | 97.75% | 2.32s | 16.1s | 15 tps | 164K | $0.259 / $0.420 | |
| Degraded | 95.34% | 95.13% | 95.98% | 97.84% | 934ms | 6.97s | 40 tps | 164K | $0.260 / $0.380 | |
| Degraded | 94.73% | 90.21% | 92.21% | 97.06% | 702ms | 10.9s | 41 tps | 131K | $0.370 / $1.11 | |
| Down | 87.35% | 90.52% | 92.50% | 98.08% | 1.36s | 5.01s | 38 tps | 164K | $0.209 / $0.310 | |
Maramara | No data | — | 95.98% | 84.62% | 86.88% | 3.42s | 6.25s | 28 tps | 33K | $3.00 / $4.50 |
| No data | — | — | 99.98% | 99.96% | — | — | — | 164K | $0.269 / $0.400 | |
SambaNovasambanova | No data | — | 96.63% | 95.49% | 96.14% | 2.65s | 8.45s | 44 tps | 33K | $3.00 / $4.50 |
| No data | — | — | — | 99.49% | — | — | — | 128K | $0.214 / $0.322 |
- Baidu100%Healthy24h 98.74%
- 7d
- 96.02%
- 30d
- 98.94%
- p50
- 1.49s
- p99
- 5.18s
- Throughput
- 43 tps
- Context
- 131K
- $/Mtok
- $0.280 / $0.420
baidu/fp8fp8 - DeepInfra100%Healthy24h 99.74%
- 7d
- 99.09%
- 30d
- 99.05%
- p50
- 1.51s
- p99
- 10.6s
- Throughput
- 15 tps
- Context
- 164K
- $/Mtok
- $0.260 / $0.380
deepinfra/fp4fp4 - Friendli100%Healthy24h 99.97%
- 7d
- 99.89%
- 30d
- 99.96%
- p50
- 1.03s
- p99
- 8.25s
- Throughput
- 43 tps
- Context
- 164K
- $/Mtok
- $0.500 / $1.50
friendli - Google100%Healthy24h 99.12%
- 7d
- 99.52%
- 30d
- 99.18%
- p50
- 1.70s
- p99
- 23.0s
- Throughput
- 37 tps
- Context
- 164K
- $/Mtok
- $0.560 / $1.68
google-vertex - Phala100%Healthy24h 99.57%
- 7d
- 99.56%
- 30d
- 99.42%
- p50
- 2.97s
- p99
- 13.1s
- Throughput
- 11 tps
- Context
- 164K
- $/Mtok
- $1.00 / $1.00
phala - Venice100%Healthy24h 99.79%
- 7d
- 98.83%
- 30d
- 98.83%
- p50
- 1.70s
- p99
- 7.71s
- Throughput
- 14 tps
- Context
- 160K
- $/Mtok
- $0.268 / $0.390
venice - DigitalOcean99.97%Healthy24h 98.04%
- 7d
- 97.56%
- 30d
- 97.81%
- p50
- 915ms
- p99
- 7.50s
- Throughput
- 32 tps
- Context
- 164K
- $/Mtok
- $0.300 / $0.960
digitalocean - SiliconFlow97.15%Healthy24h 95.73%
- 7d
- 96.20%
- 30d
- 97.75%
- p50
- 2.32s
- p99
- 16.1s
- Throughput
- 15 tps
- Context
- 164K
- $/Mtok
- $0.259 / $0.420
siliconflow/fp8fp8 - AtlasCloud95.34%Degraded24h 95.13%
- 7d
- 95.98%
- 30d
- 97.84%
- p50
- 934ms
- p99
- 6.97s
- Throughput
- 40 tps
- Context
- 164K
- $/Mtok
- $0.260 / $0.380
atlas-cloud/fp8fp8 - Alibaba94.73%Degraded24h 90.21%
- 7d
- 92.21%
- 30d
- 97.06%
- p50
- 702ms
- p99
- 10.9s
- Throughput
- 41 tps
- Context
- 131K
- $/Mtok
- $0.370 / $1.11
alibaba/fp8fp8 - GMICloud87.35%Down24h 90.52%
- 7d
- 92.50%
- 30d
- 98.08%
- p50
- 1.36s
- p99
- 5.01s
- Throughput
- 38 tps
- Context
- 164K
- $/Mtok
- $0.209 / $0.310
gmicloud/fp8fp8 - Mara—No data24h 95.98%
- 7d
- 84.62%
- 30d
- 86.88%
- p50
- 3.42s
- p99
- 6.25s
- Throughput
- 28 tps
- Context
- 33K
- $/Mtok
- $3.00 / $4.50
mara - No data24h —
- 7d
- 99.98%
- 30d
- 99.96%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 164K
- $/Mtok
- $0.269 / $0.400
novita/fp8fp8 - No data24h 96.63%
- 7d
- 95.49%
- 30d
- 96.14%
- p50
- 2.65s
- p99
- 8.45s
- Throughput
- 44 tps
- Context
- 33K
- $/Mtok
- $3.00 / $4.50
sambanova - No data24h —
- 7d
- —
- 30d
- 99.49%
- p50
- —
- p99
- —
- Throughput
- —
- Context
- 128K
- $/Mtok
- $0.214 / $0.322
streamlake/fp8fp8
Measured 5m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
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.
Daily availability
One row per endpoint · one dot per day · last 90 days.
worst hour ≥ 95%90–95%< 90%no data
Colour by each endpoint's worst hour of each UTC day · 15 endpoints · 48 days with data, 148 endpoint-days below 90%.
Price changes
- DeepSeek V3.2AtlasCloudatlas-cloud/fp8$0.134 → $0.260 / $0.202 → $0.380cache read $0.067 → $0.1302026-09-30 23:10 UTC
- DeepSeek V3.2AtlasCloudatlas-cloud/fp8$0.260 → $0.134 / $0.380 → $0.202cache read $0.130 → $0.0672026-09-28 17:45 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.290 → $0.209 / $0.430 → $0.310cache read $0.030 → $0.0222026-09-26 00:50 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.209 → $0.290 / $0.310 → $0.430cache read $0.022 → $0.0302026-09-26 00:40 UTC
- DeepSeek V3.2DigitalOceandigitalocean$0.250 → $0.300 / $0.800 → $0.960cache read $0.075 → $0.0902026-09-14 15:15 UTC
- DeepSeek V3.2Venicevenice$0.260 → $0.268 / $0.378 → $0.390cache read $0.126 → $0.1302026-09-12 18:51 UTC
- DeepSeek V3.2Venicevenice$0.330 → $0.260 / $0.480 → $0.378cache read $0.160 → $0.1262026-09-12 18:11 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.290 → $0.209 / $0.430 → $0.310cache read $0.030 → $0.0222026-09-09 10:10 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.209 → $0.290 / $0.310 → $0.430cache read $0.022 → $0.0302026-09-09 10:05 UTC
- DeepSeek V3.2DigitalOceandigitalocean$0.500 → $0.250 / $1.60 → $0.800cache read $0.150 → $0.0752026-08-28 22:05 UTC
- DeepSeek V3.2DigitalOceandigitalocean$0.250 → $0.500 / $0.800 → $1.60cache read $0.075 → $0.1502026-08-27 22:50 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.290 → $0.209 / $0.430 → $0.310cache read $0.030 → $0.0222026-08-27 18:45 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.209 → $0.290 / $0.310 → $0.430cache read $0.022 → $0.0302026-08-27 18:20 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.290 → $0.209 / $0.430 → $0.310cache read $0.030 → $0.0222026-08-26 18:30 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.209 → $0.290 / $0.310 → $0.430cache read $0.022 → $0.0302026-08-26 14:35 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.290 → $0.209 / $0.430 → $0.310cache read $0.030 → $0.0222026-08-26 13:50 UTC
- DeepSeek V3.2GMICloudgmicloud/fp8$0.209 → $0.290 / $0.310 → $0.430cache read $0.022 → $0.0302026-08-26 13:46 UTC