DeepSeek V3
deepseek/deepseek-chat
Compare with DeepSeek V4 Flash 0731, DeepSeek V4 Flash 0423, DeepSeek V4 Pro 0423
CompareDeepSeek V3
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 100% | 96.43% | 98.51% | 98.51% | 1.23s | 4.16s | 18 tps | 64K | $0.400 / $1.30 | |
StreamLakestreamlake | Healthy | 98.63% | 98.99% | 99.22% | 99.22% | 1.46s | 7.03s | 18 tps | 128K | $0.257 / $1.03 |
| Healthy | 98.51% | 88.62% | 91.57% | 91.57% | 536ms | 5.81s | 11 tps | 164K | $0.320 / $0.890 |
- Novita100%Healthy24h 96.43%
- 7d
- 98.51%
- 30d
- 98.51%
- p50
- 1.23s
- p99
- 4.16s
- Throughput
- 18 tps
- Context
- 64K
- $/Mtok
- $0.400 / $1.30
novita/fp8fp8 - StreamLake98.63%Healthy24h 98.99%
- 7d
- 99.22%
- 30d
- 99.22%
- p50
- 1.46s
- p99
- 7.03s
- Throughput
- 18 tps
- Context
- 128K
- $/Mtok
- $0.257 / $1.03
streamlake - DeepInfra98.51%Healthy24h 88.62%
- 7d
- 91.57%
- 30d
- 91.57%
- p50
- 536ms
- p99
- 5.81s
- Throughput
- 11 tps
- Context
- 164K
- $/Mtok
- $0.320 / $0.890
deepinfra/fp4fp4
Measured 3m 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.
Throughput
Median tokens per second.