Skip to content

Benchmark

AHELM: MuTox - EM

Scores are percentages.

Models measured
12
Scores recorded
359
Sources
1
IQ weight
No weight published

Score distribution

75.5%
Mean
77.9%
Median
15.6%
Std dev
53.4%
10th percentile
100%
90th percentile
33.3%
Min
33.3%100%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Gemini 2.0 Flashlanguage: Indonesian· AHELM (HELM Audio) data100%
2Gemini 2.0 Flash Experimentallanguage: Indonesian· AHELM (HELM Audio) data100%
3Gemini 2.0 Flash Litelanguage: Dutch· AHELM (HELM Audio) data100%
4Gemini 2.0 Flash Litelanguage: French· AHELM (HELM Audio) data100%
5Gemini 2.0 Flash Litelanguage: Indonesian· AHELM (HELM Audio) data100%
6Gemini 2.0 Flash Litelanguage: Tagalog· AHELM (HELM Audio) data100%
7Gemini 2.5 Flash Preview 05-20 (no think)language: French· AHELM (HELM Audio) data100%
8Gemini 2.5 Flash Preview 05-20 (no think)language: Indonesian· AHELM (HELM Audio) data100%
9Gemini 2.5 Pro Preview 05-06language: Indonesian· AHELM (HELM Audio) data100%
10gemini-1.5-flash-001language: French· AHELM (HELM Audio) data100%
11gemini-1.5-flash-002language: French· AHELM (HELM Audio) data100%
12gemini-1.5-flash-002language: Hindi· AHELM (HELM Audio) data100%
13gemini-1.5-flash-002language: Indonesian· AHELM (HELM Audio) data100%
14gemini-1.5-pro-001language: French· AHELM (HELM Audio) data100%
15gemini-1.5-pro-001language: Indonesian· AHELM (HELM Audio) data100%
16gemini-1.5-pro-001language: Tagalog· AHELM (HELM Audio) data100%
17gemini-1.5-pro-001language: Urdu· AHELM (HELM Audio) data100%
18gemini-1.5-pro-002language: Dutch· AHELM (HELM Audio) data100%
19gemini-1.5-pro-002language: French· AHELM (HELM Audio) data100%
20gemini-1.5-pro-002language: Hindi· AHELM (HELM Audio) data100%
21gemini-1.5-pro-002language: Indonesian· AHELM (HELM Audio) data100%
22gemini-1.5-pro-002language: Tagalog· AHELM (HELM Audio) data100%
23gemini-1.5-pro-002language: Urdu· AHELM (HELM Audio) data100%
24GPT-4o Audio Previewlanguage: French· AHELM (HELM Audio) data100%
25GPT-4o Audio Previewlanguage: Hindi· AHELM (HELM Audio) data100%
26GPT-4o Audio Previewlanguage: Indonesian· AHELM (HELM Audio) data100%
27GPT-4o Audio Previewlanguage: Turkish· AHELM (HELM Audio) data100%
28GPT-4o Audio Previewlanguage: Urdu· AHELM (HELM Audio) data100%
29GPT-4o mini Audio Previewlanguage: Dutch· AHELM (HELM Audio) data100%
30GPT-4o mini Audio Previewlanguage: French· AHELM (HELM Audio) data100%
31GPT-4o mini Audio Previewlanguage: Hindi· AHELM (HELM Audio) data100%
32GPT-4o mini Audio Previewlanguage: Indonesian· AHELM (HELM Audio) data100%
33GPT-4o mini Audio Previewlanguage: Tagalog· AHELM (HELM Audio) data100%
34GPT-4o mini Audio Previewlanguage: Turkish· AHELM (HELM Audio) data100%
35GPT-4o mini Audio Previewlanguage: Urdu· AHELM (HELM Audio) data100%
36Qwen2.5-Omni 7Blanguage: Bengali· AHELM (HELM Audio) data100%
37Qwen2.5-Omni 7Blanguage: French· AHELM (HELM Audio) data100%
38Qwen2.5-Omni 7Blanguage: Indonesian· AHELM (HELM Audio) data100%
39GPT-4o Audio Previewlanguage: Estonian· AHELM (HELM Audio) data94.6%
40GPT-4o Audio Previewlanguage: Hebrew· AHELM (HELM Audio) data94.1%
41GPT-4o Audio Previewlanguage: Catalan· AHELM (HELM Audio) data92.4%
42Gemini 2.5 Flash Preview 05-20 (no think)language: Dutch· AHELM (HELM Audio) data92.3%
43gemini-1.5-flash-002language: Dutch· AHELM (HELM Audio) data92.3%
44gemini-1.5-pro-001language: Dutch· AHELM (HELM Audio) data92.3%
45GPT-4o Audio Previewlanguage: Dutch· AHELM (HELM Audio) data92.3%
46Qwen2.5-Omni 7Blanguage: Dutch· AHELM (HELM Audio) data92.3%
47GPT-4o Audio Previewlanguage: Slovak· AHELM (HELM Audio) data91.9%
48gemini-1.5-pro-002language: Catalan· AHELM (HELM Audio) data91.9%
49GPT-4o mini Audio Previewlanguage: Catalan· AHELM (HELM Audio) data91.9%
50GPT-4o mini Audio Previewlanguage: Estonian· AHELM (HELM Audio) data91.6%

Showing 1–50 of 359 results.