HyperAI
Command Palette
Search for a command to run...
AA-LCR v1.1: Long Context Reasoning
言語モデルが10kから100kトークン（cl100k_baseトークナイザで計測）に及ぶ長文文書から情報を抽出し、推論し、統合する能力を測定する難易度の高いベンチマーク。
- モデル
- 259 現在、 596 合計
- 中央値
- 61.3%
- 範囲
- 0.0%–88.3%
| ランク | 比較する | モデル | スコア |
|---|---|---|---|
| 51 |  GPT-6 Astra (Low) OpenAI | 80.0% | |
| 52 | GPT-6 Astra (Xhigh) OpenAI | 80.0% | |
| 53 | GPT-6 Luna (Xhigh) OpenAI | 80.0% | |
| 54 |  Grok 4.7 (Low) SpaceXAI | 80.0% | |
| 55 |  K2 Horizon 375B A23B Institute of Foundation Models | 80.0% | |
| 56 | 79.7% | ||
| 57 |  Gemini 4 Argon (High) Google | 79.7% | |
| 58 |  GLM-5.3 (Max) Z AI | 79.7% | |
| 59 | GPT-6.1 Sol (Xhigh) OpenAI | 79.7% | |
| 60 | GPT-6 Astra (Medium) OpenAI | 79.7% | |
| 61 |  MiniMax-M3.1-Flash MiniMax | 79.7% | |
| 62 |  Qwen3.8 27B (Medium) Alibaba | 79.7% | |
| 63 | Qwen3.8-Flash-Next Alibaba | 79.7% | |
| 64 |  Apodex 1.1 Apodex | 79.3% | |
| 65 | Gemini 4 Carbon (High) Google | 79.3% | |
| 66 | GPT-6 Luna (High) OpenAI | 79.3% | |
| 67 |  Kimi K2.7 Code Kimi | 79.3% | |
| 68 | 79.3% | ||
| 69 |  Hy3 Tencent | 79.0% | |
| 70 |  Claude Haiku 5.5 (Xhigh) Anthropic | 78.3% | |
| 71 | GPT-6 Luna (Medium) OpenAI | 78.3% | |
| 72 |  JT-4.1 Flash 236B A21B (Reasoning) China Mobile | 78.3% | |
| 73 |  Ling-3.0-flash-VL InclusionAI | 78.3% | |
| 74 | 78.3% | ||
| 75 | 78.0% | ||
| 76 |  Beam 1.0 Reflection | 78.0% | |
| 77 | Claude Haiku 5.5 (High) Anthropic | 77.3% | |
| 78 | Claude Haiku 5.5 (Medium) Anthropic | 77.3% | |
| 79 |  Inkling (Xhigh) Thinking Machines | 77.3% | |
| 80 | 77.3% | ||
| 81 | Qwen3.8 27B (Low) Alibaba | 77.3% | |
| 82 | Grok 4.7 (High) SpaceXAI | 77.0% | |
| 83 | Grok 4.7 (Xhigh) SpaceXAI | 76.7% | |
| 84 | 76.3% | ||
| 85 |  Quasar 438B (Max, Based on GLM-5.2) Multiverse Computing | 76.3% | |
| 86 | 76.3% | ||
| 87 | 76.0% | ||
| 88 | Gemini 3.5 Flash-Lite Google | 76.0% | |
| 89 |  Motif 3 Motif Technologies | 76.0% | |
| 90 | Inkling Small Thinking Machines | 75.7% | |
| 91 | o3 OpenAI | 74.7% | |
| 92 |  MiMo-V2.6-Flash Xiaomi | 74.3% | |
| 93 | GPT-6 Luna (Low) OpenAI | 74.0% | |
| 94 | Ling-3.0-flash-Fin InclusionAI | 73.7% | |
| 95 |  KAT Coder Pro V2 KwaiKAT | 73.0% | |
| 96 | Qwen3.7 Plus Alibaba | 73.0% | |
| 97 | GLM-5.3 (Low) Z AI | 72.7% | |
| 98 | Claude Haiku 5.5 (Low) Anthropic | 72.0% | |
| 99 |  Doubao Seed Code ByteDance Seed | 71.7% | |
| 100 | K2 Horizon MoVA 36B A4B Institute of Foundation Models | 71.7% |
