Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Open-ended Instruction Following on AlpacaEval Audio
Loading...
4.59
EN Score
SEA-LION-v3-8B-IT
3.5292
3.8046
4.08
4.3554
Mar 7, 2026
EN Score
ZH Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
EN Score
ZH Score
SEA-LION-v3-8B-IT
Approach category=Text...
2026.03
4.59
4.21
MERaLiON-2-10B
Approach category=Zero...
2026.03
4.59
1.96
Ours (soft-gating)
Approach category=Dist...
2026.03
4.58
3.21
Ours (hard-gating)
Approach category=Dist...
2026.03
4.39
3.33
ML-DiVA
Approach category=Dist...
2026.03
4.28
2.87
Whisper-v3 + SEA-LION-v3-8B-IT
Approach category=Casc...
2026.03
3.99
3.95
SeaLLMs-Audio
Approach category=Zero...
2026.03
3.93
3.72
Qwen2-Audio
Approach category=Zero...
2026.03
3.77
3.65
Glm-4-voice
Approach category=Zero...
2026.03
3.62
3.41
EN-DiVA
Approach category=Dist...
2026.03
3.57
-
Feedback
Search any
task
Search any
task