Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #33: Academic dishonesty

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
1.17

Score distribution

73.8%
Mean
80%
Median
25.8%
Std dev
33.3%
10th percentile
100%
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 1.17 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 docs66.7%
52DeepSeek R1· Stanford HELM specialized suites data docs60%
53GPT 4osetting not named by the source· Stanford HELM specialized suites data docs60%
54Llama 3.1 405B Instruct Turbo· Stanford HELM specialized suites data docs58.3%
55Yi 34B Chat· Stanford HELM specialized suites data docs56.7%
56GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs53.3%
57Grok 3 Mini· Stanford HELM specialized suites data docs53.3%
58Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs53.3%
59Grok 3· Stanford HELM specialized suites data docs51.7%
60Llama 4 Scout· Stanford HELM specialized suites data docs48.3%
61Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs48.3%
62O1 Mini· Stanford HELM specialized suites data docs46.7%
63DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs40%
64Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs40%
65Mistral Large2407· Stanford HELM specialized suites data docs38.3%
66Grok 4· Stanford HELM specialized suites data docs36.7%
67Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs36.7%
68GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs33.3%
69DeepSeek V3· Stanford HELM specialized suites data docs30%
70Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs30%
71Mistral Small· Stanford HELM specialized suites data docs26.7%
72Command R+· Stanford HELM specialized suites data docs23.3%
73Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs18.3%
74Command R Plus· Stanford HELM specialized suites data docs15%
75DBRX-instruct· 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