Configured providers
3
Ordered routing targets
Top Models
All time
Non-streamed
deepseek/deepseek-v4-flash
47.4%
qwen/qwen3.6-35b-a3b
42.1%
qwen/qwen3.8-27b
10.5%
Streamed
deepseek/deepseek-v4-flash
52.9%
qwen/qwen3.6-35b-a3b
37.5%
gpt-5-nano-2025-08-07
7.5%
qwen/qwen3.8-27b
1.2%
qwen/qwen3.5-35b-a3b
0.3%
deepseek-ai/DeepSeek-V3.1
0.3%
openai/gpt-5-nano
0.3%
lod/auto
0.1%
lod-paperclip
0.0%
Last 24
Non-streamed
No model usage data yet.
Streamed
gpt-5-nano-2025-08-07
100.0%
Failover Rate
All time
Non-streamed
1st attempt
52.6%
2nd attempt
47.4%
Streamed
2nd attempt
52.8%
1st attempt
39.8%
3rd attempt
7.4%
Last 24
Non-streamed
No failover success data yet.
Streamed
3rd attempt
100.0%
Average Tokens / Second by Model
All time
Non-streamed
| Model | Avg tokens/sec |
|---|---|
| qwen/qwen3.6-35b-a3b | 99.70 |
| deepseek/deepseek-v4-flash | 29.70 |
| qwen/qwen3.8-27b | 5.86 |
Streamed
| Model | Avg tokens/sec |
|---|---|
| gpt-5-nano-2025-08-07 | 131.06 |
| deepseek/deepseek-v4-flash | 46.57 |
| qwen/qwen3.8-27b | 27.97 |
Last 24
Non-streamed
No token throughput data yet.
Streamed
No token throughput data yet.
Failed Requests
All time
Non-streamed
Failed request rate
32.14%
9 of 28 requests
Streamed
Failed request rate
0.02%
6 of 26034 requests
Last 24
Non-streamed
No request data yet.
Streamed
Failed request rate
0.00%
0 of 50 requests