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
301Gemini 2.0 Flashlanguage: Bulgarian· AHELM (HELM Audio) data58.3%
302Gemini 2.0 Flashlanguage: Polish· AHELM (HELM Audio) data58%
303Gemini 2.0 Flashlanguage: Hindi· AHELM (HELM Audio) data57.1%
304Gemini 2.0 Flashlanguage: Turkish· AHELM (HELM Audio) data57.1%
305Gemini 2.0 Flash Experimentallanguage: Hindi· AHELM (HELM Audio) data57.1%
306Gemini 2.0 Flash Experimentallanguage: Turkish· AHELM (HELM Audio) data57.1%
307Gemini 2.5 Pro Preview 05-06language: Hindi· AHELM (HELM Audio) data57.1%
308Gemini 2.5 Pro Preview 05-06language: Vietnamese· AHELM (HELM Audio) data57.1%
309Gemini 2.0 Flash Experimentallanguage: Polish· AHELM (HELM Audio) data56.8%
310Gemini 2.0 Flashlanguage: Italian· AHELM (HELM Audio) data56.3%
311Gemini 2.0 Flash Experimentallanguage: Italian· AHELM (HELM Audio) data56.3%
312Gemini 2.0 Flashlanguage: Mandarin_Chinese· AHELM (HELM Audio) data55.6%
313Gemini 2.0 Flash Experimentallanguage: Mandarin_Chinese· AHELM (HELM Audio) data55.6%
314Gemini 2.5 Pro Preview 05-06language: Russian· AHELM (HELM Audio) data55.6%
315Gemini 2.0 Flash Experimentallanguage: Finnish· AHELM (HELM Audio) data54.6%
316gemini-1.5-flash-001language: Bulgarian· AHELM (HELM Audio) data54.3%
317Gemini 2.0 Flashlanguage: Finnish· AHELM (HELM Audio) data54%
318gemini-1.5-flash-001language: Polish· AHELM (HELM Audio) data53.8%
319Qwen2.5-Omni 7Blanguage: English· AHELM (HELM Audio) data53.8%
320Gemini 2.0 Flashlanguage: Slovak· AHELM (HELM Audio) data53.8%
321Gemini 2.0 Flash Experimentallanguage: Slovak· AHELM (HELM Audio) data53.8%
322gemini-1.5-pro-001language: English· AHELM (HELM Audio) data53.5%
323gemini-1.5-flash-001language: Greek· AHELM (HELM Audio) data53.4%
324Gemini 2.0 Flashlanguage: Catalan· AHELM (HELM Audio) data53%
325Gemini 2.0 Flash Experimentallanguage: Catalan· AHELM (HELM Audio) data53%
326gemini-1.5-flash-002language: Spanish· AHELM (HELM Audio) data50.9%
327gemini-1.5-flash-001language: Catalan· AHELM (HELM Audio) data50.8%
328Gemini 2.0 Flashlanguage: Czech· AHELM (HELM Audio) data50.4%
329Gemini 2.0 Flash Experimentallanguage: Czech· AHELM (HELM Audio) data50.4%
330Gemini 2.0 Flashlanguage: Vietnamese· AHELM (HELM Audio) data50%
331Gemini 2.0 Flash Experimentallanguage: Vietnamese· AHELM (HELM Audio) data50%
332Gemini 2.5 Pro Preview 05-06language: Portuguese· AHELM (HELM Audio) data50%
333gemini-1.5-flash-001language: Arabic· AHELM (HELM Audio) data50%
334gemini-1.5-flash-001language: Italian· AHELM (HELM Audio) data50%
335gemini-1.5-flash-001language: Portuguese· AHELM (HELM Audio) data50%
336gemini-1.5-flash-002language: Hebrew· AHELM (HELM Audio) data49.3%
337gemini-1.5-flash-001language: Hungarian· AHELM (HELM Audio) data49.2%
338Gemini 2.0 Flashlanguage: Hebrew· AHELM (HELM Audio) data47.8%
339Gemini 2.0 Flashlanguage: Estonian· AHELM (HELM Audio) data47.6%
340Gemini 2.0 Flashlanguage: Greek· AHELM (HELM Audio) data47.3%
341Gemini 2.0 Flash Experimentallanguage: Greek· AHELM (HELM Audio) data47.3%
342Gemini 2.0 Flash Experimentallanguage: Estonian· AHELM (HELM Audio) data47%
343Gemini 2.0 Flash Experimentallanguage: Hebrew· AHELM (HELM Audio) data46.8%
344gemini-1.5-flash-001language: Czech· AHELM (HELM Audio) data46%
345Gemini 2.0 Flashlanguage: English· AHELM (HELM Audio) data45.8%
346Gemini 2.0 Flash Experimentallanguage: English· AHELM (HELM Audio) data45.8%
347gemini-1.5-flash-001language: Slovak· AHELM (HELM Audio) data45.7%
348gemini-1.5-flash-001language: Spanish· AHELM (HELM Audio) data45.4%
349gemini-1.5-flash-001language: Estonian· AHELM (HELM Audio) data45.2%
350gemini-1.5-flash-001language: Russian· AHELM (HELM Audio) data44.4%

Showing 301–350 of 359 results.