Skip to content

Benchmark

Horangi 4: ALT_환각방지

Higher scores are better.

Models measured
79
Scores recorded
83
Sources
1
IQ weight
1.67

Score distribution

0.604
Mean
0.63
Median
0.115
Std dev
0.44
10th percentile
0.715
90th percentile
0.277
Min
0.2770.87
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.67 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
51Mistral Medium 3.5mistral-medium-3-5· Horangi Korean Leaderboard 4 data docs0.59
52GPT-5 nanogpt-5-nano-2025-08-07· Horangi Korean Leaderboard 4 data docs0.587
53GPT OSS 120Bgpt-oss-120b: high-effort· Horangi Korean Leaderboard 4 data docs0.583
54GPT-5.4 minigpt-5.4-mini-2026-03-17: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.58
55Grok 4.1 Fastgrok-4-1-fast-non-reasoning· Horangi Korean Leaderboard 4 data docs0.58
56Claude Haiku 4.5claude-haiku-4-5-20251001: high-effort· Horangi Korean Leaderboard 4 data docs0.577
57GPT-4.1 nanogpt-4.1-nano· Horangi Korean Leaderboard 4 data docs0.577
58Qwen3 14BQwen3-14B· Horangi Korean Leaderboard 4 data docs0.563
59GPT-5.6 Lunagpt-5.6-luna: max-effort· Horangi Korean Leaderboard 4 data docs0.557
60GPT-4.1 minigpt-4.1-mini· Horangi Korean Leaderboard 4 data docs0.553
61Grok 4.20grok-4.20· Horangi Korean Leaderboard 4 data docs0.547
62Nemotron 3 Nano 30B A3BNVIDIA-Nemotron-3-Nano-30B-A3B-BF16· Horangi Korean Leaderboard 4 data docs0.54
63GLM 4.5 Airglm-4.5-air· Horangi Korean Leaderboard 4 data docs0.539
64Mistral Smallmistral-small-2603· Horangi Korean Leaderboard 4 data docs0.537
65MiMo V2 Flashmimo-v2-flash· Horangi Korean Leaderboard 4 data docs0.536
66Qwen3 4B Instruct 2507Qwen3-4B-Instruct-2507· Horangi Korean Leaderboard 4 data docs0.533
67Grok 4.20grok-4.20-beta· Horangi Korean Leaderboard 4 data docs0.52
68Gemma 4 E4B-itgemma-4-E4B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.511
69GLM 4.7 Flashglm-4.7-flash· Horangi Korean Leaderboard 4 data docs0.511
70Qwen3 8BQwen3-8B· Horangi Korean Leaderboard 4 data docs0.51
71Grok 4 Fastgrok-4-fast-non-reasoning· Horangi Korean Leaderboard 4 data docs0.48
72MiniMax M2.5minimax-m2.5· Horangi Korean Leaderboard 4 data docs0.476
73Qwen3 4BQwen3-4B· Horangi Korean Leaderboard 4 data docs0.473
74Gemma 4 E2B-itgemma-4-E2B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.454
75Qwen3 1.7BQwen3-1.7B· Horangi Korean Leaderboard 4 data docs0.44
76GPT 5.4 Nanogpt-5.4-nano-2026-03-17: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.433
77Qwen3 0.6BQwen3-0.6B· Horangi Korean Leaderboard 4 data docs0.433
78MiniMax M2.7minimax-m2.7· Horangi Korean Leaderboard 4 data docs0.43
79Gemma 3 27B ITgemma-3-27b-it· Horangi Korean Leaderboard 4 data docs0.367
80Gemma 3 1b itgemma-3-1b-it· Horangi Korean Leaderboard 4 data docs0.343
81Gemma 3 12B ITgemma-3-12b-it· Horangi Korean Leaderboard 4 data docs0.33
82Gemma 3 270M Itgemma-3-270m-it· Horangi Korean Leaderboard 4 data docs0.29
83Gemma 3 4B ITgemma-3-4b-it· Horangi Korean Leaderboard 4 data docs0.277

Showing 51–83 of 83 results.

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