Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #2: Integrity

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
1.59

Score distribution

78.9%
Mean
88.9%
Median
24.1%
Std dev
34.6%
10th percentile
100%
90th percentile
9.9%
Min
9.9%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.59 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
51gemini-2.5-flash-preview-04-17· Stanford HELM specialized suites data docs76.5%
52GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs75.9%
53Qwen2.5 72B Instruct Turbo· Stanford HELM specialized suites data docs75.9%
54Llama 3.1 405B Instruct Turbo· Stanford HELM specialized suites data docs74.1%
55Grok 3· Stanford HELM specialized suites data docs71%
56DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs66%
57Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs64.2%
58Grok 3 Mini· Stanford HELM specialized suites data docs63%
59Llama 4 Scout· Stanford HELM specialized suites data docs62.3%
60DeepSeek V3· Stanford HELM specialized suites data docs58%
61Yi 34B Chat· Stanford HELM specialized suites data docs58%
62Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs56.8%
63Grok 4· Stanford HELM specialized suites data docs53.1%
64Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs53.1%
65Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs48.1%
66Mistral Large2407· Stanford HELM specialized suites data docs46.9%
67DeepSeek R1· Stanford HELM specialized suites data docs46.3%
68GLM 4.5 Air· Stanford HELM specialized suites data docs34.6%
69Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs30.9%
70Command R+· Stanford HELM specialized suites data docs28.4%
71Mistral Small· Stanford HELM specialized suites data docs28.4%
72O1 Mini· Stanford HELM specialized suites data docs25.9%
73Command R Plus· Stanford HELM specialized suites data docs21.6%
74Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs14.2%
75DBRX-instruct· Stanford HELM specialized suites data docs9.9%

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