Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - Violence & Extremism

Safety benchmarks examine risks in model behaviour, including how models respond to unsafe requests. Scores are percentages. Higher is better.

Models measured
70
Scores recorded
75
Sources
1
IQ weight
0.94

Score distribution

61.4%
Mean
63.3%
Median
20.6%
Std dev
29.7%
10th percentile
84.8%
90th percentile
16.1%
Min
16.1%94.3%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.94 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-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs55%
52GLM 4.5 Air· Stanford HELM specialized suites data docs53.4%
53Llama 4 Scout· Stanford HELM specialized suites data docs49.2%
54GPT 4osetting not named by the source· Stanford HELM specialized suites data docs48.9%
55DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs46.7%
56Yi 34B Chat· Stanford HELM specialized suites data docs44.9%
57Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs43.9%
58GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs42.8%
59Grok 3· Stanford HELM specialized suites data docs42.1%
60Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs42%
61Grok 3 Mini· Stanford HELM specialized suites data docs41.2%
62DeepSeek R1· Stanford HELM specialized suites data docs37.5%
63O1 Mini· Stanford HELM specialized suites data docs36.3%
64Grok 4· Stanford HELM specialized suites data docs35.4%
65Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs35.4%
66GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs35%
67DeepSeek V3· Stanford HELM specialized suites data docs31.7%
68Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs29.7%
69Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs27.9%
70Mistral Large2407· Stanford HELM specialized suites data docs27.4%
71Mistral Small· Stanford HELM specialized suites data docs24.7%
72Command R+· Stanford HELM specialized suites data docs24.1%
73Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs19.9%
74Command R Plus· Stanford HELM specialized suites data docs19.6%
75DBRX-instruct· Stanford HELM specialized suites data docs16.1%

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