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
251GPT-4o Audio Previewlanguage: English· AHELM (HELM Audio) data69.1%
252gemini-1.5-flash-002language: Italian· AHELM (HELM Audio) data68.8%
253gemini-1.5-flash-002language: Danish· AHELM (HELM Audio) data68.4%
254gemini-1.5-flash-002language: Bulgarian· AHELM (HELM Audio) data68.3%
255gemini-1.5-flash-002language: Czech· AHELM (HELM Audio) data68.1%
256gemini-1.5-pro-001language: Czech· AHELM (HELM Audio) data68.1%
257GPT-4o mini Audio Previewlanguage: Spanish· AHELM (HELM Audio) data68%
258GPT-4o mini Audio Previewlanguage: English· AHELM (HELM Audio) data67.9%
259gemini-1.5-pro-001language: Hungarian· AHELM (HELM Audio) data67.7%
260gemini-1.5-flash-002language: Polish· AHELM (HELM Audio) data67.5%
261Gemini 2.0 Flashlanguage: Danish· AHELM (HELM Audio) data67.3%
262gemini-1.5-pro-001language: Hebrew· AHELM (HELM Audio) data67%
263Gemini 2.0 Flashlanguage: Russian· AHELM (HELM Audio) data66.7%
264Gemini 2.0 Flash Experimentallanguage: Danish· AHELM (HELM Audio) data66.7%
265Gemini 2.0 Flash Experimentallanguage: Russian· AHELM (HELM Audio) data66.7%
266gemini-1.5-flash-002language: Hungarian· AHELM (HELM Audio) data66.7%
267Qwen2.5-Omni 7Blanguage: Portuguese· AHELM (HELM Audio) data66.7%
268gemini-1.5-pro-002language: Spanish· AHELM (HELM Audio) data66.5%
269gemini-1.5-flash-002language: Finnish· AHELM (HELM Audio) data66.3%
270Gemini 2.0 Flash Litelanguage: Vietnamese· AHELM (HELM Audio) data64.3%
271GPT-4o Audio Previewlanguage: German· AHELM (HELM Audio) data64.3%
272GPT-4o Audio Previewlanguage: Vietnamese· AHELM (HELM Audio) data64.3%
273Gemini 2.0 Flash Litelanguage: English· AHELM (HELM Audio) data64.1%
274Gemini 2.5 Flash Preview 05-20 (no think)language: Spanish· AHELM (HELM Audio) data64%
275Gemini 2.5 Flash Preview 05-20 (no think)language: English· AHELM (HELM Audio) data63.9%
276Gemini 2.0 Flash Litelanguage: Spanish· AHELM (HELM Audio) data63.3%
277Qwen2.5-Omni 7Blanguage: Spanish· AHELM (HELM Audio) data63%
278gemini-1.5-flash-002language: Catalan· AHELM (HELM Audio) data62.7%
279Gemini 2.5 Flash Preview 05-20 (no think)language: Italian· AHELM (HELM Audio) data62.5%
280Gemini 2.5 Pro Preview 05-06language: French· AHELM (HELM Audio) data62.5%
281Gemini 2.5 Pro Preview 05-06language: Italian· AHELM (HELM Audio) data62.5%
282gemini-1.5-pro-001language: Italian· AHELM (HELM Audio) data62.5%
283gemini-1.5-pro-002language: Italian· AHELM (HELM Audio) data62.5%
284Gemini 2.0 Flash· AHELM (HELM Audio) data62.1%
285Gemini 2.0 Flash Experimental· AHELM (HELM Audio) data62%
286Gemini 2.5 Pro Preview 05-06language: Spanish· AHELM (HELM Audio) data61%
287Gemini 2.5 Pro Preview 05-06language: Swahili· AHELM (HELM Audio) data60%
288gemini-1.5-flash-001language: Swahili· AHELM (HELM Audio) data60%
289gemini-1.5-flash-002language: Arabic· AHELM (HELM Audio) data60%
290gemini-1.5-pro-001language: Arabic· AHELM (HELM Audio) data60%
291gemini-1.5-pro-001language: Spanish· AHELM (HELM Audio) data60%
292Gemini 2.5 Pro Preview 05-06language: English· AHELM (HELM Audio) data59.8%
293gemini-1.5-pro-002language: English· AHELM (HELM Audio) data59.4%
294gemini-1.5-flash-001· AHELM (HELM Audio) data59.1%
295Gemini 2.0 Flash Experimentallanguage: Bulgarian· AHELM (HELM Audio) data58.8%
296gemini-1.5-flash-001language: Danish· AHELM (HELM Audio) data58.5%
297Gemini 2.0 Flashlanguage: Portuguese· AHELM (HELM Audio) data58.3%
298Gemini 2.0 Flash Experimentallanguage: Portuguese· AHELM (HELM Audio) data58.3%
299Gemini 2.0 Flash Litelanguage: Portuguese· AHELM (HELM Audio) data58.3%
300Gemini 2.5 Flash Preview 05-20 (no think)language: Portuguese· AHELM (HELM Audio) data58.3%

Showing 251–300 of 359 results.