Skip to content

Benchmark

Horangi 4 subset: haerae_bench_v1_rc

Higher scores are better.

Models measured
79
Scores recorded
83
Sources
1
IQ weight
2.46

Score distribution

0.844
Mean
0.88
Median
0.12
Std dev
0.76
10th percentile
0.92
90th percentile
0.2
Min
0.20.93
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.46 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 docs0.93
2Gemini 3.5 Flashgemini-3.5-flash· Horangi Korean Leaderboard 4 data docs0.93
3GPT-5.6 Lunagpt-5.6-luna: max-effort· Horangi Korean Leaderboard 4 data docs0.93
4GPT-5.6 Solgpt-5.6-sol: max-effort· Horangi Korean Leaderboard 4 data docs0.93
5Claude Opus 4.6claude-opus-4-6: high-effort· Horangi Korean Leaderboard 4 data docs0.92
6Claude Opus 4.8claude-opus-4-8: high-effort· Horangi Korean Leaderboard 4 data docs0.92
7Claude Sonnet 4.6claude-sonnet-4-6: high-effort· Horangi Korean Leaderboard 4 data docs0.92
8GPT 5.2gpt-5.2-2025-12-11: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.92
9GPT 5.5gpt-5.5: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.92
10Grok 4.1 Fastgrok-4-1-fast-reasoning· Horangi Korean Leaderboard 4 data docs0.92
11Claude Opus 4.5claude-opus-4-5-20251101: high-effort· Horangi Korean Leaderboard 4 data docs0.91
12Claude Sonnet 4.5claude-sonnet-4-5-20250929: high-effort· Horangi Korean Leaderboard 4 data docs0.91
13Claude Sonnet 5claude-sonnet-5: high-effort· Horangi Korean Leaderboard 4 data docs0.91
14Gemini 3 Pro Previewgemini-3-pro-preview: high-effort· Horangi Korean Leaderboard 4 data docs0.91
15Qwen3.5 397B A17Bqwen3.5-397b-a17b· Horangi Korean Leaderboard 4 data docs0.91
16Qwen3.6 Flashqwen3.6-flash· Horangi Korean Leaderboard 4 data docs0.91
17Qwen 3 235B A22B Thinking 2507qwen3-235b-a22b-thinking-2507· Horangi Korean Leaderboard 4 data docs0.909
18Qwen3.7 Maxqwen3.7-max· Horangi Korean Leaderboard 4 data docs0.909
19Claude Opus 4.1claude-opus-4-1-20250805· Horangi Korean Leaderboard 4 data docs0.9
20Gemini 3.1 Flash Litegemini-3.1-flash-lite-preview· Horangi Korean Leaderboard 4 data docs0.9
21Gemini 3.1 Pro Previewgemini-3.1-pro-preview: high-effort· Horangi Korean Leaderboard 4 data docs0.9
22Gemma 4 31B ITGemma-4-31B-IT-NVFP4: enable-thinking· Horangi Korean Leaderboard 4 data docs0.9
23GPT 5.4gpt-5.4-2026-03-05: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.9
24GPT-5.6 Terragpt-5.6-terra: max-effort· Horangi Korean Leaderboard 4 data docs0.9
25Grok 4 Fastgrok-4-fast-reasoning· Horangi Korean Leaderboard 4 data docs0.9
26Grok 4.20grok-4.20· Horangi Korean Leaderboard 4 data docs0.9
27Grok 4.5grok-4.5· Horangi Korean Leaderboard 4 data docs0.9
28Qwen3.5 27Bqwen3.5-27b· Horangi Korean Leaderboard 4 data docs0.9
29Qwen3.7 Plusqwen3.7-plus· Horangi Korean Leaderboard 4 data docs0.9
30GLM 5 Turboglm-5-turbo· Horangi Korean Leaderboard 4 data docs0.899
31Qwen3.6 Plus Previewqwen3.6-plus-preview· Horangi Korean Leaderboard 4 data docs0.898
32Gemma 4 31B ITgemma-4-31B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.89
33GLM 5glm-5· Horangi Korean Leaderboard 4 data docs0.89
34GLM 5.2glm-5.2: high-effort· Horangi Korean Leaderboard 4 data docs0.89
35GLM 5V Turboglm-5v-turbo· Horangi Korean Leaderboard 4 data docs0.89
36GPT 5gpt-5-2025-08-07· Horangi Korean Leaderboard 4 data docs0.89
37GPT-5.4 minigpt-5.4-mini-2026-03-17: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.89
38MiniMax M2.7minimax-m2.7· Horangi Korean Leaderboard 4 data docs0.89
39Qwen3.5 122B-A10Bqwen3.5-122b-a10b· Horangi Korean Leaderboard 4 data docs0.89
40Kimi K2.5kimi-k2.5· Horangi Korean Leaderboard 4 data docs0.888
41Qwen3.5 9Bqwen3.5-9b· Horangi Korean Leaderboard 4 data docs0.882
42Claude Opus 4claude-opus-4-20250514: high-effort· Horangi Korean Leaderboard 4 data docs0.88
43GPT 5.1gpt-5.1-2025-11-13· Horangi Korean Leaderboard 4 data docs0.88
44GPT 5.4 Nanogpt-5.4-nano-2026-03-17: xhigh-effort· Horangi Korean Leaderboard 4 data docs0.88
45MiMo V2 Promimo-v2-pro· Horangi Korean Leaderboard 4 data docs0.88
46Qwen3.5 35B A3Bqwen3.5-35b-a3b· Horangi Korean Leaderboard 4 data docs0.88
47Grok 4.20grok-4.20-beta· Horangi Korean Leaderboard 4 data docs0.879
48Claude Haiku 4.5claude-haiku-4-5-20251001: high-effort· Horangi Korean Leaderboard 4 data docs0.87
49DeepSeek Reasonerdeepseek-v3.2: high-effort· Horangi Korean Leaderboard 4 data docs0.87
50Gemma 4 26B A4Bgemma-4-26B-A4B-it: enable-thinking· Horangi Korean Leaderboard 4 data docs0.87

Showing 1–50 of 83 results.

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