DeepSeek V3.1
deepseek/deepseek-chat-v3.1
Compare with DeepSeek V3.1 Terminus, DeepSeek V3, DeepSeek V3.2
CompareDeepSeek V3.1
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 99.90% | 99.86% | 99.86% | 1.61s | 3.54s | 33 tps | 131K | $0.300 / $0.950 | |
| Healthy | 100% | 99.99% | 99.99% | 99.99% | 355ms | 997ms | 45 tps | 161K | $0.550 / $1.65 | |
Googlegoogle-vertex/us-west2 | Healthy | 100% | 95.04% | 95.32% | 95.32% | 1.24s | 2.85s | 71 tps | 164K | $0.600 / $1.70 |
| Healthy | 100% | 99.99% | 99.99% | 99.99% | 1.77s | 8.76s | 21 tps | 131K | $0.270 / $1.00 | |
| Healthy | 100% | 96.81% | 97.60% | 97.60% | 1.24s | 8.68s | 34 tps | 131K | $0.650 / $1.50 | |
| Healthy | 100% | 97.92% | 98.02% | 98.02% | 1.34s | 10.0s | 18 tps | 164K | $0.270 / $1.00 | |
| Healthy | 99.67% | 99.88% | 99.50% | 99.50% | 745ms | 5.73s | 13 tps | 164K | $0.250 / $0.950 | |
Maramara | No data | — | 93.76% | 81.37% | 81.37% | 1.62s | 11.1s | 70 tps | 131K | $0.600 / $1.70 |
- AtlasCloud100%Healthy24h 99.90%
- 7d
- 99.86%
- 30d
- 99.86%
- p50
- 1.61s
- p99
- 3.54s
- Throughput
- 33 tps
- Context
- 131K
- $/Mtok
- $0.300 / $0.950
atlas-cloud/fp8fp8 - CoreWeave100%Healthy24h 99.99%
- 7d
- 99.99%
- 30d
- 99.99%
- p50
- 355ms
- p99
- 997ms
- Throughput
- 45 tps
- Context
- 161K
- $/Mtok
- $0.550 / $1.65
coreweave/fp8fp8 - Google100%Healthy24h 95.04%
- 7d
- 95.32%
- 30d
- 95.32%
- p50
- 1.24s
- p99
- 2.85s
- Throughput
- 71 tps
- Context
- 164K
- $/Mtok
- $0.600 / $1.70
google-vertex/us-west2 - Novita100%Healthy24h 99.99%
- 7d
- 99.99%
- 30d
- 99.99%
- p50
- 1.77s
- p99
- 8.76s
- Throughput
- 21 tps
- Context
- 131K
- $/Mtok
- $0.270 / $1.00
novita/fp8fp8 - SambaNova100%Healthy24h 96.81%
- 7d
- 97.60%
- 30d
- 97.60%
- p50
- 1.24s
- p99
- 8.68s
- Throughput
- 34 tps
- Context
- 131K
- $/Mtok
- $0.650 / $1.50
sambanova/fp8fp8 - SiliconFlow100%Healthy24h 97.92%
- 7d
- 98.02%
- 30d
- 98.02%
- p50
- 1.34s
- p99
- 10.0s
- Throughput
- 18 tps
- Context
- 164K
- $/Mtok
- $0.270 / $1.00
siliconflow/fp8fp8 - DeepInfra99.67%Healthy24h 99.88%
- 7d
- 99.50%
- 30d
- 99.50%
- p50
- 745ms
- p99
- 5.73s
- Throughput
- 13 tps
- Context
- 164K
- $/Mtok
- $0.250 / $0.950
deepinfra/fp4fp4 - Mara—No data24h 93.76%
- 7d
- 81.37%
- 30d
- 81.37%
- p50
- 1.62s
- p99
- 11.1s
- Throughput
- 70 tps
- Context
- 131K
- $/Mtok
- $0.600 / $1.70
mara
Measured 1m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
1 endpoint reporting no data is 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 is OpenRouter's rolling 30-minute window.
Throughput
Median tokens per second.