router.huggingface.co

API Provider Performance Test Results

Host Addressrouter.huggingface.co
Supported Models7 models
Supported Models
7 models
Model NameTest CountAverage Output SpeedAverage First Token Latency
moonshotai/Kimi-K2-Instruct:novita
5
48.81 t/s
1.21s
openai/gpt-oss-120b
5
225.67 t/s
0.92s
openai/gpt-oss-120b:novita
5
240.32 t/s
1.26s
openai/gpt-oss-20b:novita
5
155.67 t/s
3.56s
Qwen/Qwen3-235B-A22B:novita
5
34.12 t/s
1.12s
Qwen/Qwen3-Coder-480B-A35B-Instruct:novita
5
63.59 t/s
1.06s
zai-org/GLM-4.5:novita
5
35.36 t/s
1.39s
Total Tests
35

Total Tests

Average Tokens/Second
114.79 t/s

26.37 - 265.38 t/s

Average First Token Latency
1.50s

0.63 - 8.91s

Last Test Time
8/19/2025, 7:25:23 AM

Last Test Time

Top Models Performance
1

openai/gpt-oss-120b:novita

Test Count: 5

240.32 t/s

Average Speed

2

openai/gpt-oss-120b

Test Count: 5

225.67 t/s

Average Speed

3

openai/gpt-oss-20b:novita

Test Count: 5

155.67 t/s

Average Speed

4

Qwen/Qwen3-Coder-480B-A35B-Instruct:novita

Test Count: 5

63.59 t/s

Average Speed

5

moonshotai/Kimi-K2-Instruct:novita

Test Count: 5

48.81 t/s

Average Speed

Recent Test Records
7 records
Test TimeModelAverage Output SpeedAverage First Token LatencyTotal Tokens
8/19/2025, 7:25:23 AM
openai/gpt-oss-120b
225.67 t/s
0.92s
5630
8/13/2025, 3:29:37 PM
moonshotai/Kimi-K2-Instruct:novita
48.81 t/s
1.21s
2423
8/13/2025, 3:26:54 PM
Qwen/Qwen3-Coder-480B-A35B-Instruct:novita
63.59 t/s
1.06s
1772
8/13/2025, 2:45:43 PM
openai/gpt-oss-20b:novita
155.67 t/s
3.56s
4701
8/13/2025, 2:41:25 PM
openai/gpt-oss-120b:novita
240.32 t/s
1.26s
7011
8/13/2025, 2:35:13 PM
zai-org/GLM-4.5:novita
35.36 t/s
1.39s
5178
8/13/2025, 3:46:09 AM
Qwen/Qwen3-235B-A22B:novita
34.12 t/s
1.12s
3072