OpenInference
3 endpoints · all healthy · worst uptime 99.76%
| State | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Healthy | 99.95% | 99.48% | 97.38% | 94.64% | 4.11s | 25.2s | 15 tps | 1.0M | $0.046 / $0.650 | |
| Healthy | 99.78% | 99.85% | 98.44% | 98.44% | 1.50s | 8.45s | 31 tps | 1.0M | $0.020 / $1.66 | |
| Healthy | 99.76% | 99.52% | 98.24% | 98.24% | 1.09s | 9.26s | 33 tps | 1.0M | $0.029 / $1.66 |
- GLM 5.3 Flash99.95%Healthy24h 99.48%
- 7d
- 97.38%
- 30d
- 94.64%
- p50
- 4.11s
- p99
- 25.2s
- Throughput
- 15 tps
- Context
- 1.0M
- $/Mtok
- $0.046 / $0.650
open-inference/fp4fp4 - DeepSeek V4 Flash 073199.78%Healthy24h 99.85%
- 7d
- 98.44%
- 30d
- 98.44%
- p50
- 1.50s
- p99
- 8.45s
- Throughput
- 31 tps
- Context
- 1.0M
- $/Mtok
- $0.020 / $1.66
open-inference/fp4fp4 - DeepSeek V4 Flash 042399.76%Healthy24h 99.52%
- 7d
- 98.24%
- 30d
- 98.24%
- p50
- 1.09s
- p99
- 9.26s
- Throughput
- 33 tps
- Context
- 1.0M
- $/Mtok
- $0.029 / $1.66
open-inference/fp4fp4
Measured 2m ago. Latency and throughput are OpenRouter's rolling 30-minute windows.
History
GLM 5.3 Flash open-inference/fp4
DeepSeek V4 Flash 0731 open-inference/fp4
DeepSeek V4 Flash 0423 open-inference/fp4
Colour by daily mean uptime:≥ 95%90–95%< 90%no data
Through the live last bucket (raw rounds, updated every collection round) · 3 endpoints with data, sorted worst-first.