DeepSeek V3.1 Terminus: 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) | 99.91% | 99.83% | 99.83% | 1.76s | 5.40s | 37 tps◆ (best) | 131K | $0.300 / $0.950 | |
| Healthy | 100%◆ (best) | 99.99% | 99.99%◆ (best) | 99.99%◆ (best) | 2.19s | 10.8s | 19 tps | 131K | $0.270 / $1.00◆ (best) | |
| Healthy | 100%◆ (best) | 97.66% | 98.03% | 98.03% | 1.76s | 13.3s | 20 tps | 164K | $0.270 / $1.00◆ (best) | |
StreamLakestreamlake | No data | — | 99.82% | 99.86% | 99.86% | 1.47s◆ (best) | 5.36s◆ (best) | 35 tps | 128K | $0.343 / $1.03 |
- AtlasCloud100%◆ (best)Healthy24h 99.91%
- 7d
- 99.83%
- 30d
- 99.83%
- p50
- 1.76s
- p99
- 5.40s
- Throughput
- 37 tps◆ (best)
- Context
- 131K
- $/Mtok
- $0.300 / $0.950
atlas-cloud/fp8fp8 - Novita100%◆ (best)Healthy24h 99.99%
- 7d
- 99.99%◆ (best)
- 30d
- 99.99%◆ (best)
- p50
- 2.19s
- p99
- 10.8s
- Throughput
- 19 tps
- Context
- 131K
- $/Mtok
- $0.270 / $1.00◆ (best)
novita/fp8fp8 - SiliconFlow100%◆ (best)Healthy24h 97.66%
- 7d
- 98.03%
- 30d
- 98.03%
- p50
- 1.76s
- p99
- 13.3s
- Throughput
- 20 tps
- Context
- 164K
- $/Mtok
- $0.270 / $1.00◆ (best)
siliconflow/fp8fp8 - No data24h 99.82%
- 7d
- 99.86%
- 30d
- 99.86%
- p50
- 1.47s◆ (best)
- p99
- 5.36s◆ (best)
- Throughput
- 35 tps
- Context
- 128K
- $/Mtok
- $0.343 / $1.03
streamlake
Measured 4m 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.
Throughput
Median tokens per second.