Benchmark
Nejumi 4: ALT_堅牢性
Higher scores are better.
Benchmark sourceScores from Nejumi Leaderboard 4 data docs.
- Models measured
- 114
- Scores recorded
- 126
- Sources
- 1
- IQ weight
- 2.00
Score distribution
- 0.826
- Mean
- 0.89
- Median
- 0.159
- Std dev
- 0.6
- 10th percentile
- 0.97
- 90th percentile
- 0.31
- Min
This benchmark has a weight of 2.00 for chat models, measured over 114 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.
Leaderboard
| # | Model | Score |
|---|---|---|
| 51 | o3 Miniopenai/o3-mini-2025-01-31· Nejumi Leaderboard 4 data docs | 0.91 |
| 52 | Claude Sonnet 4anthropic/claude-sonnet-4-20250514: extended-thinking· Nejumi Leaderboard 4 data docs | 0.905 |
| 53 | Gemini 3.1 Flash Litegoogle/gemini-3.1-flash-lite-preview· Nejumi Leaderboard 4 data docs | 0.905 |
| 54 | Grok 3 Minigrok-3-mini· Nejumi Leaderboard 4 data docs | 0.905 |
| 55 | o4 Miniopenai/o4-mini-2025-04-16· Nejumi Leaderboard 4 data docs | 0.905 |
| 56 | Qwen3 Next 80B A3BQwen/Qwen3-Next-80B-A3B-Thinking: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.905 |
| 57 | Claude Opus 4.5anthropic/claude-opus-4.5-20251125: extended-thinking· Nejumi Leaderboard 4 data docs | 0.9 |
| 58 | DeepSeek Reasonerdeepseek/DeepSeek-V3.2 (Thinking Mode)· Nejumi Leaderboard 4 data docs | 0.9 |
| 59 | Kimi K2 Thinkingmoonshotai/kimi-k2-thinking· Nejumi Leaderboard 4 data docs | 0.9 |
| 60 | Qwen3 Next 80B A3BQwen/Qwen3-Next-80B-A3B-Thinking· Nejumi Leaderboard 4 data docs | 0.895 |
| 61 | DeepSeek V4 Prodeepseek-ai/DeepSeek-V4-Pro: reasoning-max· Nejumi Leaderboard 4 data docs | 0.89 |
| 62 | Kimi K2.5moonshotai/Kimi-K2.5: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.89 |
| 63 | Qwen3 32BQwen/Qwen3-32B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.89 |
| 64 | Claude Opus 4.5anthropic/claude-opus-4.5-20251125: no-thinking· Nejumi Leaderboard 4 data docs | 0.88 |
| 65 | GPT-5.4 miniopenai/gpt-5.4-mini-2026-03-17: high-effort· Nejumi Leaderboard 4 data docs | 0.88 |
| 66 | DeepSeek V3.2deepseek-ai/DeepSeek-V3.2: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.875 |
| 67 | MiniMax M3MiniMaxAI/MiniMax-M3: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.875 |
| 68 | Qwen3 30B A3BQwen/Qwen3-30B-A3B-Thinking-2507: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.87 |
| 69 | Claude Haiku 4.5anthropic/claude-haiku-4-5-20251001: extended-thinking· Nejumi Leaderboard 4 data docs | 0.865 |
| 70 | Qwen3-4B-Thinking-2507Qwen/Qwen3-4B-Thinking-2507· Nejumi Leaderboard 4 data docs | 0.86 |
| 71 | Qwen3-4B-Thinking-2507Qwen/Qwen3-4B-Thinking-2507: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.86 |
| 72 | Qwen3.5 4BQwen/Qwen3.5-4B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.855 |
| 73 | Claude 3.5 Sonnetanthropic/claude-3.5-sonnet-20241022· Nejumi Leaderboard 4 data docs | 0.85 |
| 74 | Claude Sonnet 3.7anthropic/claude-3.7-sonnet-20250219: extended-thinking· Nejumi Leaderboard 4 data docs | 0.85 |
| 75 | Qwen 3 235B A22BQwen/Qwen3-235B-A22B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.85 |
| 76 | DeepSeek V4 Prodeepseek/deepseek-v4-pro: thinking-max· Nejumi Leaderboard 4 data docs | 0.845 |
| 77 | Gemini 2.5 Flashgoogle/gemini-2.5-flash· Nejumi Leaderboard 4 data docs | 0.845 |
| 78 | MiniMax M2.1MiniMaxAI/MiniMax-M2.1: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.84 |
| 79 | Qwen3 14BQwen/Qwen3-14B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.84 |
| 80 | Claude Sonnet 4anthropic/claude-sonnet-4-20250514: no-thinking· Nejumi Leaderboard 4 data docs | 0.835 |
| 81 | Grok 3grok-3· Nejumi Leaderboard 4 data docs | 0.835 |
| 82 | Qwen: Qwen3 VL 8B ThinkingQwen/Qwen3-VL-8B-Thinking· Nejumi Leaderboard 4 data docs | 0.83 |
| 83 | Qwen3 Max PreviewQwen/Qwen3-Max-Preview· Nejumi Leaderboard 4 data docs | 0.83 |
| 84 | Qwen3 4BQwen/Qwen3-4B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.82 |
| 85 | GPT OSS 120Bopenai/gpt-oss-120b: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.815 |
| 86 | GPT 4.1openai/gpt-4-1-2025-04-14· Nejumi Leaderboard 4 data docs | 0.805 |
| 87 | GPT 5.1openai/gpt-5.1-2025-11-13: none-effort· Nejumi Leaderboard 4 data docs | 0.805 |
| 88 | GPT-5 nanoopenai/gpt-5-nano-2025-08-07: high-effort· Nejumi Leaderboard 4 data docs | 0.805 |
| 89 | Claude Sonnet 3.7anthropic/claude-3.7-sonnet-20250219: no-thinking· Nejumi Leaderboard 4 data docs | 0.8 |
| 90 | Qwen3 8BQwen/Qwen3-8B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.795 |
| 91 | Qwen3 30B A3BQwen/Qwen3-30B-A3B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.79 |
| 92 | Qwen3 Next 80B A3B InstructQwen/Qwen3-Next-80B-A3B-Instruct· Nejumi Leaderboard 4 data docs | 0.79 |
| 93 | DeepSeek R1deepseek-ai/DeepSeek-R1-0528: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.785 |
| 94 | QwQ 32BQwen/QwQ-32B: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.78 |
| 95 | GPT 4oopenai/gpt-4o-2024-11-20· Nejumi Leaderboard 4 data docs | 0.77 |
| 96 | MiniMax M2MiniMaxAI/MiniMax-M2: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.77 |
| 97 | GLM 4.7 Flashzai-org/GLM-4.7-Flash: reasoning-enabled· Nejumi Leaderboard 4 data docs | 0.74 |
| 98 | DeepSeek V3deepseek-ai/DeepSeek-V3-0324· Nejumi Leaderboard 4 data docs | 0.735 |
| 99 | Llama 3.1 405B Instructmeta-llama/Llama-3.1-405B-Instruct-FP8· Nejumi Leaderboard 4 data docs | 0.7 |
| 100 | Claude Haiku 3.5anthropic/claude-3.5-haiku-20241022· Nejumi Leaderboard 4 data docs | 0.695 |
Showing 51–100 of 126 results.
Data credits: Nejumi Leaderboard 4, Weights & Biases Japan; public publisher-selected numeric evaluation results