Skip to content

Benchmark

Horangi 4 subset: korean_hate_speech

Higher scores are better.

Models measured
79
Scores recorded
83
Sources
1
IQ weight
0.31

Score distribution

0.607
Mean
0.616
Median
0.065
Std dev
0.53
10th percentile
0.69
90th percentile
0.455
Min
0.4550.78
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.31 for chat models, measured over 79 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Claude Fable 5claude-fable-5: high-effort· Horangi Korean Leaderboard 4 data docs0.78
2Claude Haiku 4.5claude-haiku-4-5-20251001: high-effort· Horangi Korean Leaderboard 4 data docs0.76
3Gemma 3 27B ITgemma-3-27b-it· Horangi Korean Leaderboard 4 data docs0.74
4Claude Opus 4claude-opus-4-20250514: high-effort· Horangi Korean Leaderboard 4 data docs0.71
5DeepSeek Reasonerdeepseek-v3.2: high-effort· Horangi Korean Leaderboard 4 data docs0.7
6Grok 4.1 Fastgrok-4-1-fast-non-reasoning· Horangi Korean Leaderboard 4 data docs0.7
7Claude Opus 4.6claude-opus-4-6: high-effort· Horangi Korean Leaderboard 4 data docs0.69
8GLM 5 Turboglm-5-turbo· Horangi Korean Leaderboard 4 data docs0.69
9Grok 4.20grok-4.20· Horangi Korean Leaderboard 4 data docs0.69
10Mistral Medium 3.5mistral-medium-3-5· Horangi Korean Leaderboard 4 data docs0.69
11Grok 4.20grok-4.20-beta· Horangi Korean Leaderboard 4 data docs0.687
12Claude Sonnet 4.6claude-sonnet-4-6: high-effort· Horangi Korean Leaderboard 4 data docs0.67
13Mistral Smallmistral-small-2603· Horangi Korean Leaderboard 4 data docs0.67
14Qwen3.5 397B A17Bqwen3.5-397b-a17b· Horangi Korean Leaderboard 4 data docs0.67
15Claude Opus 4.1claude-opus-4-1-20250805· Horangi Korean Leaderboard 4 data docs0.66
16Claude Sonnet 4.5claude-sonnet-4-5-20250929: high-effort· Horangi Korean Leaderboard 4 data docs0.66
17Gemma 3 12B ITgemma-3-12b-it· Horangi Korean Leaderboard 4 data docs0.66
18GLM 5V Turboglm-5v-turbo· Horangi Korean Leaderboard 4 data docs0.66
19Grok 4 Fastgrok-4-fast-non-reasoning· Horangi Korean Leaderboard 4 data docs0.66
20MiMo V2 Promimo-v2-pro· Horangi Korean Leaderboard 4 data docs0.66
21Qwen3.7 Maxqwen3.7-max· Horangi Korean Leaderboard 4 data docs0.66
22Claude Opus 4.5claude-opus-4-5-20251101: high-effort· Horangi Korean Leaderboard 4 data docs0.65
23GLM 4.7glm-4.7· Horangi Korean Leaderboard 4 data docs0.65
24Grok 4.1 Fastgrok-4-1-fast-reasoning· Horangi Korean Leaderboard 4 data docs0.64
25MiMo V2 Flashmimo-v2-flash· Horangi Korean Leaderboard 4 data docs0.64
26Qwen3.6 Plus Previewqwen3.6-plus-preview· Horangi Korean Leaderboard 4 data docs0.64
27GLM 5.2glm-5.2: high-effort· Horangi Korean Leaderboard 4 data docs0.636
28Kimi K2.5kimi-k2.5· Horangi Korean Leaderboard 4 data docs0.636
29Gemini 3 Flashgemini-3-flash-preview· Horangi Korean Leaderboard 4 data docs0.63
30Gemma 4 26B A4Bgemma-4-26B-A4B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.63
31GPT 4.1gpt-4.1· Horangi Korean Leaderboard 4 data docs0.63
32GPT 4ogpt-4o-2024-11-20· Horangi Korean Leaderboard 4 data docs0.63
33Grok 4 Fastgrok-4-fast-reasoning· Horangi Korean Leaderboard 4 data docs0.63
34Grok 4.5grok-4.5· Horangi Korean Leaderboard 4 data docs0.63
35Qwen3 4B Instruct 2507Qwen3-4B-Instruct-2507· Horangi Korean Leaderboard 4 data docs0.63
36Qwen3.5 35B A3Bqwen3.5-35b-a3b· Horangi Korean Leaderboard 4 data docs0.63
37Claude Opus 4.8claude-opus-4-8: high-effort· Horangi Korean Leaderboard 4 data docs0.62
38Gemma 4 31B ITgemma-4-31B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.62
39Qwen3.5 27Bqwen3.5-27b· Horangi Korean Leaderboard 4 data docs0.62
40Qwen3.7 Plusqwen3.7-plus· Horangi Korean Leaderboard 4 data docs0.62
41GLM 4.5 Airglm-4.5-air· Horangi Korean Leaderboard 4 data docs0.616
42GLM 5glm-5· Horangi Korean Leaderboard 4 data docs0.616
43Claude Sonnet 5claude-sonnet-5: high-effort· Horangi Korean Leaderboard 4 data docs0.61
44Gemini 3.1 Pro Previewgemini-3.1-pro-preview: high-effort· Horangi Korean Leaderboard 4 data docs0.61
45Gemma 4 31B ITGemma-4-31B-IT-NVFP4: enable-thinking· Horangi Korean Leaderboard 4 data docs0.61
46Qwen3 0.6BQwen3-0.6B· Horangi Korean Leaderboard 4 data docs0.6
47Qwen3 1.7BQwen3-1.7B· Horangi Korean Leaderboard 4 data docs0.6
48Qwen3 4BQwen3-4B· Horangi Korean Leaderboard 4 data docs0.6
49Gemma 3 4B ITgemma-3-4b-it· Horangi Korean Leaderboard 4 data docs0.59
50GPT-4.1 minigpt-4.1-mini· Horangi Korean Leaderboard 4 data docs0.59

Showing 1–50 of 83 results.

Data credits: Horangi 4, Weights & Biases Korea; public publisher-selected numeric evaluation results