Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Output Diversity on INFINITY-EVAL
Loading...
39.61
Distinct Score
DeepSeek-14B
19.6316
24.8183
30.005
35.1917
Jan 16, 2026
Distinct Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Distinct Score
DeepSeek-14B
sampling_setting=S-best
2026.01
39.61
DeepSeek-14B
sampling_setting=Mixed
2026.01
35.33
DeepSeek-14B
sampling_setting=S-non...
2026.01
31.84
Qwen3-32B
sampling_setting=Mixed
2026.01
31.47
Qwen3-32B
sampling_setting=S-best
2026.01
28.66
Qwen3-8B
sampling_setting=Mixed
2026.01
28.13
Qwen3-32B
sampling_setting=S-non...
2026.01
27.52
Qwen3-14B
sampling_setting=S-best
2026.01
27.07
Qwen3-32B
sampling_setting=S-en
2026.01
27
Qwen3-14B
sampling_setting=Mixed
2026.01
26.73
DeepSeek-14B
sampling_setting=S-en
2026.01
25.27
Qwen3-8B
sampling_setting=S-best
2026.01
24.51
Qwen3-14B
sampling_setting=S-non...
2026.01
22.6
Qwen3-8B
sampling_setting=S-non...
2026.01
22.54
Qwen3-8B
sampling_setting=S-en
2026.01
20.67
Qwen3-14B
sampling_setting=S-en
2026.01
20.4
Feedback
Search any
task
Search any
task