Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #20: Monetized

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
0.61

Score distribution

38.6%
Mean
26.7%
Median
25.9%
Std dev
13.3%
10th percentile
81.7%
90th percentile
10%
Min
10%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.61 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
51DBRX-instruct· Stanford HELM specialized suites data docs18.3%
52Grok 3· Stanford HELM specialized suites data docs18.3%
53Grok 3 Mini· Stanford HELM specialized suites data docs18.3%
54Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs18.3%
55O1 Mini· Stanford HELM specialized suites data docs18.3%
56Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs18.3%
57GPT 4.1· Stanford HELM specialized suites data docs16.7%
58GPT 4osetting not named by the source· Stanford HELM specialized suites data docs16.7%
59Grok 4· Stanford HELM specialized suites data docs16.7%
60Command R Plus· Stanford HELM specialized suites data docs15%
61GPT 4· Stanford HELM specialized suites data docs15%
62GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs15%
63GPT-4o mini· Stanford HELM specialized suites data docs15%
64GPT-5 nano· Stanford HELM specialized suites data docs15%
65Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs15%
66GPT-4.1 mini· Stanford HELM specialized suites data docs13.3%
67Mistral Large2407· Stanford HELM specialized suites data docs13.3%
68Mistral Small· Stanford HELM specialized suites data docs13.3%
69o1· Stanford HELM specialized suites data docs13.3%
70DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs11.7%
71DeepSeek V3· Stanford HELM specialized suites data docs11.7%
72Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs11.7%
73Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs11.7%
74o3 Mini· Stanford HELM specialized suites data docs11.7%
75Command R+· Stanford HELM specialized suites data docs10%

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