Skip to content

Benchmark

HELM AIR-Bench: AIRBench 2024 - #35: Sowing Division

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
1.66

Score distribution

75.7%
Mean
86.7%
Median
26.9%
Std dev
30%
10th percentile
100%
90th percentile
3.3%
Min
3.3%100%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.66 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 docs73.3%
52GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs73.3%
53GPT 4osetting not named by the source· Stanford HELM specialized suites data docs70%
54GPT 4osetting not named by the source· Stanford HELM specialized suites data docs66.7%
55Llama 4 Scout· Stanford HELM specialized suites data docs66.7%
56DeepSeek LLM 67B Chat· Stanford HELM specialized suites data docs60%
57DeepSeek R1· Stanford HELM specialized suites data docs60%
58Qwen2.5 7B Instruct Turbo· Stanford HELM specialized suites data docs60%
59GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs53.3%
60Qwen1.5-72B-Chat· Stanford HELM specialized suites data docs53.3%
61Grok 3 Mini· Stanford HELM specialized suites data docs50%
62Grok 3· Stanford HELM specialized suites data docs41.7%
63Meta Llama 3.1 70B Instruct Turbo· Stanford HELM specialized suites data docs41.7%
64GPT-3.5 Turbosetting not named by the source· Stanford HELM specialized suites data docs40%
65Mixtral 8x22B Instruct v0.1· Stanford HELM specialized suites data docs40%
66Mixtral 8x7B Instruct v0.1· Stanford HELM specialized suites data docs36.7%
67Grok 4· Stanford HELM specialized suites data docs33.3%
68DeepSeek V3· Stanford HELM specialized suites data docs30%
69O1 Mini· Stanford HELM specialized suites data docs26.7%
70Command R Plus· Stanford HELM specialized suites data docs23.3%
71Command R+· Stanford HELM specialized suites data docs20%
72Mistral Large2407· Stanford HELM specialized suites data docs20%
73Mistral (7B) Instruct v0.3· Stanford HELM specialized suites data docs13.3%
74Mistral Small· Stanford HELM specialized suites data docs13.3%
75DBRX-instruct· Stanford HELM specialized suites data docs3.3%

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