HyperAI
Command Palette
Search for a command to run...
HLE-Diamond
Humanity's Last Examの洗練・更新された1,000問のセット。推論問題と知識問題が均等に分かれ、272問が画像を含む。
- カテゴリ
- 一般
- インテリジェンス指標の重み
- 10%
- モデル
- 133 現在、 162 合計
- 中央値
- 17.3%
- 範囲
- 2.8%–80.2%
| ランク | 比較する | モデル | スコア | コスト / タスク | 時間 / タスク | 出力 token / タスク |
|---|---|---|---|---|---|---|
| 1 | 80.2% | $1.73 | 7m 1s | 43.2K | ||
| 2 | 75.8% | $0.88 | 3m 31s | 21.8K | ||
| 3 | 73.8% | $0.46 | 2m 3s | 11.5K | ||
| 4 | Gemini 4 Carbon (High) Google | 69.1% | $1.63 | 13m 47s | 68K | |
| 5 | 67.9% | $1.20 | 8m 31s | 59.7K | ||
| 6 | Gemini 4 Carbon (Medium) Google | 67.4% | $0.79 | 7m 39s | 33.9K | |
| 7 | 66.3% | $0.18 | 48.9s | 4.44K | ||
| 8 | 64.0% | $0.47 | 3m 35s | 23.3K | ||
| 9 | GPT-6 Astra (Max) OpenAI | 63.8% | $0.55 | 4m 50s | 10.9K | |
| 10 | Gemini 4 Carbon (Low) Google | 61.9% | $0.69 | 7m 29s | 28.9K | |
| 11 | GPT-6 Astra (Xhigh) OpenAI | 61.5% | $0.37 | 3m 7s | 7.28K | |
| 12 | 60.8% | $0.13 | 33.1s | 3.07K | ||
| 13 | GPT-6 Astra (High) OpenAI | 59.3% | $0.26 | 2m 8s | 5.02K | |
| 14 | GPT-6.1 Sol (Max) OpenAI | 58.8% | $0.11 | 3m 12s | 10.8K | |
| 15 | Gemini 4 Argon (High) Google | 56.7% | $0.48 | 5m 5s | 20.7K | |
| 16 | GPT-6.1 Sol (Xhigh) OpenAI | 55.8% | $0.066 | 1m 57s | 6.5K | |
| 17 | 55.6% | $0.19 | 1m 31s | 9.49K | ||
| 18 | 53.3% | $0.80 | 8m 42s | 80.2K | ||
| 19 | GPT-6 Astra (Medium) OpenAI | 52.7% | $0.12 | 53.2s | 2.27K | |
| 20 | GPT-6.1 Sol (High) OpenAI | 51.9% | $0.043 | 1m 14s | 4.2K | |
| 21 | 51.4% | $0.13 | 1m 1s | 6.31K | ||
| 22 | 48.5% | — | — | 39.8K | ||
| 23 | GPT-6.1 Terra (Max) OpenAI | 47.3% | $0.15 | 6m 38s | 29.2K | |
| 24 | GPT-6.1 Sol (Medium) OpenAI | 46.0% | $0.019 | 29.3s | 1.72K | |
| 25 | GPT-6 Astra (Low) OpenAI | 45.1% | $0.05 | 21.9s | 870 | |
| 26 | 44.1% | $0.25 | 2m 56s | 24.5K | ||
| 27 | GPT-6.1 Terra (Xhigh) OpenAI | 42.3% | $0.041 | 1m 53s | 8.12K | |
| 28 | 41.7% | $0.035 | 14.7s | 1.59K | ||
| 29 | 39.0% | $0.84 | 5m 44s | 42K | ||
| 30 | GPT-6.1 Sol (Low) OpenAI | 38.2% | $0.0086 | 13.0s | 734 | |
| 31 | Kimi K3.1 Pro Kimi | 37.3% | $0.36 | 18m 40s | 36.3K | |
| 32 | GPT-6.1 Terra (High) OpenAI | 35.4% | $0.018 | 50.3s | 3.52K | |
| 33 | 34.5% | $0.42 | 2m 54s | 21.1K | ||
| 34 | 33.8% | $0.11 | 1m 15s | 10.5K | ||
| 35 | Gemini 3.8 Flash (High) Google | 33.1% | $0.098 | 3m 48s | 26.2K | |
| 36 | Tencent Hy4 Tencent | 32.7% | $0.11 | — | 44.1K | |
| 37 | Gemini 3.1 Pro Preview Google | 31.4% | — | — | — | |
| 38 | GPT-6.1 Terra (Medium) OpenAI | 31.3% | $0.009 | 24.0s | 1.66K | |
| 39 | Claude Haiku 5.5 (Max) Anthropic | 29.6% | $0.047 | 6m 7s | 93.3K | |
| 40 | 29.1% | $0.027 | 1m 3s | 7.09K | ||
| 41 | Claude Haiku 5.5 (Xhigh) Anthropic | 28.8% | $0.024 | 3m 24s | 48.3K | |
| 42 | Inkling 2 Small Thinking Machines | 28.7% | $0.053 | 6m 57s | 53.2K | |
| 43 | MiniMax-M3.1-Flash MiniMax | 28.1% | $0.019 | 3m 13s | 30.8K | |
| 44 | 26.3% | $0.12 | 2m 45s | 28.5K | ||
| 45 | DeepSeek V4.1 Flash 1017 (Max) DeepSeek | 25.3% | $0.062 | 6m 11s | 51.3K | |
| 46 | Muse Ember (Max) Meta | 25.1% | $0.024 | 1m 47s | 48.6K | |
| 47 | GPT-6.1 Terra (Low) OpenAI | 24.9% | $0.0031 | 7.4s | 499 | |
| 48 | Gemini 3.8 Flash (Low) Google | 24.3% | — | — | — | |
| 49 | Step 5 Preview StepFun | 24.3% | $0.08 | 3m 24s | 29.4K | |
| 50 | Qwen3.8 Max (0902) Alibaba | 24.0% | $0.21 | 11m 11s | 34.3K |