DeepSeek V3
deepseek/deepseek-chat
Compare with DeepSeek V4 Flash 0731, DeepSeek V4 Flash 0423, DeepSeek V4 Pro 0423
CompareDeepSeek V3
| Provider | State | Uptime 5m | 24h | p50 latency | p99 | Throughput | Context | $/Mtok |
|---|---|---|---|---|---|---|---|---|
Novitanovita/fp8fp8 | Healthy | 100% | 99.93% | 1.13s | 3.50s | 19 tps | 64K | $0.400 / $1.30 |
StreamLakestreamlake | Healthy | 99.80% | 99.46% | 1.42s | 5.14s | 28 tps | 128K | $0.257 / $1.03 |
DeepInfradeepinfra/fp4fp4 | Healthy | 99.58% | 96.13% | 431ms | 5.50s | 27 tps | 164K | $0.320 / $0.890 |
- Novita100%Healthy24h 99.93%
- p50
- 1.13s
- p99
- 3.50s
- Throughput
- 19 tps
- Context
- 64K
- $/Mtok
- $0.400 / $1.30
novita/fp8fp8 - StreamLake99.80%Healthy24h 99.46%
- p50
- 1.42s
- p99
- 5.14s
- Throughput
- 28 tps
- Context
- 128K
- $/Mtok
- $0.257 / $1.03
streamlake - DeepInfra99.58%Healthy24h 96.13%
- p50
- 431ms
- p99
- 5.50s
- Throughput
- 27 tps
- Context
- 164K
- $/Mtok
- $0.320 / $0.890
deepinfra/fp4fp4
Measured 4m 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 and p99 are OpenRouter's rolling 30-minute windows.
Throughput
Median tokens per second.