Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #14: Hate Speech (Inciting/Promoting/Expressing hatred)

Scores are percentages. Higher is better.

Models measured
70
Scores recorded
75
Sources
1
IQ weight
1.39

Score distribution

86.3%
Mean
90.6%
Median
12.7%
Std dev
67%
10th percentile
98.8%
90th percentile
50.4%
Min
50.4%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.39 for chat models, measured over 70 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
51GPT 4· Stanford HELM specialized suites data docs82.6%
52GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs81.7%
53Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs80.9%
54GPT 4osetting not named by the source· Stanford HELM specialized suites data docs80.1%
55Llama 4 Scout· Stanford HELM specialized suites data docs79.7%
56Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs79.4%
57Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs78.7%
58GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs77.2%
59DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs76.1%
60DeepSeek V3· Stanford HELM specialized suites data docs74.8%
61Grok 3· Stanford HELM specialized suites data docs73.9%
62Yi 34B Chat· Stanford HELM specialized suites data docs73.6%
63GPT-4o mini· Stanford HELM specialized suites data docs73.5%
64Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs71.6%
65Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs69%
66Mistral Large2407· Stanford HELM specialized suites data docs68.8%
67Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs68.3%
68Mistral Small· Stanford HELM specialized suites data docs67%
69Grok 4· Stanford HELM specialized suites data docs63.8%
70GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs63.5%
71Command R+· Stanford HELM specialized suites data docs59.1%
72DBRX-instruct· Stanford HELM specialized suites data docs57.8%
73O1 Mini· Stanford HELM specialized suites data docs57.5%
74GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs56.8%
75Command R Plus· Stanford HELM specialized suites data docs50.4%

Showing 51–75 of 75 results.

Data credits: Stanford HELM (Stanford CRFM); scenario scores from Capabilities, Safety, Long Context, MedHELM, AIR-Bench, MMLU and Finance