فئة AI BENCHY
ترتيب حيل مضادة للذكاء الاصطناعي
اكتشف أي نماذج الذكاء الاصطناعي تؤدي أفضل في حيل مضادة للذكاء الاصطناعي، وأيها يظل أكثر اعتمادية، وأين تظهر أكبر الفجوات. الترتيب حسب: زمن الاستجابة (المتوسط) ↓.
أسباب الإخفاق المرتبطة
| الترتيب | النموذج | الشركة | درجة حيل مضادة للذكاء الاصطناعي | متوسط الدرجة | اختبارات صحيحة | زمن الاستجابة (المتوسط) |
|---|---|---|---|---|---|---|
| #23 | Seed-2.0-Mini medium | Bytedance Seed | 7.0 | 6.9 | 2/3 | 99.0s |
| #28 | Kimi K2.5 medium | Moonshot AI | 7.0 | 6.4 | 2/3 | 85.3s |
| #24 | Qwen3.5-Flash medium | Qwen | 10.0 | 6.9 | 3/3 | 71.4s |
| #8 | Gemini 3.1 Flash Lite Preview high | 10.0 | 8.2 | 3/3 | 43.9s | |
| #34 | GPT-5 Nano medium | OpenAI | 7.0 | 5.5 | 2/3 | 37.7s |
| #18 | DeepSeek V3.2 medium | DeepSeek | 7.0 | 7.3 | 2/3 | 33.4s |
| #43 | MiniMax M2.5 medium | Minimax | 9.3 | 4.7 | 2/3 | 32.4s |
| #52 | GLM 4.7 Flash medium | Z.ai | 4.0 | 3.1 | 1/3 | 27.1s |
| #14 | GLM 5 medium | Z.ai | 10.0 | 7.4 | 3/3 | 22.3s |
| #35 | Qwen3.5-35B-A3B medium | Qwen | 10.0 | 5.5 | 3/3 | 21.8s |
| #39 | gpt-oss-120b medium | OpenAI | 7.0 | 5.1 | 2/3 | 19.8s |
| #13 | Step 3.5 Flash medium | Stepfun | 10.0 | 7.4 | 3/3 | 18.5s |
| #21 | MiMo-V2-Flash medium | Xiaomi | 9.7 | 7.2 | 3/3 | 16.8s |
| #32 | GPT-5 Mini medium | OpenAI | 7.0 | 6.0 | 2/3 | 16.5s |
| #50 | Qwen3 Coder Next medium | Qwen | 1.3 | 3.5 | 0/3 | 15.3s |
| #27 | GPT-5.2 medium | OpenAI | 7.0 | 6.5 | 2/3 | 14.3s |
| #26 | Claude Opus 4.6 medium | Anthropic | 4.0 | 6.6 | 1/3 | 11.9s |
| #46 | Kimi K2.5 none | Moonshot AI | 2.7 | 4.1 | 0/3 | 11.4s |
| #4 | Qwen3.5 Plus 2026-02-15 medium | Qwen | 10.0 | 8.3 | 3/3 | 10.4s |
| #7 | Qwen3.5-27B medium | Qwen | 10.0 | 8.2 | 3/3 | 9.69s |
| #2 | Gemini 3.1 Pro Preview medium | 10.0 | 9.4 | 3/3 | 9.52s | |
| #33 | DeepSeek V3.2 none | DeepSeek | 10.0 | 5.5 | 0/3 | 8.79s |
| #10 | Qwen3.5-122B-A10B medium | Qwen | 10.0 | 7.7 | 3/3 | 6.99s |
| #16 | Gemini 2.5 Flash medium | 7.3 | 7.4 | 2/3 | 6.98s | |
| #49 | GLM 4.7 Flash none | Z.ai | 10.0 | 3.9 | 0/3 | 6.59s |
| #30 | Grok 4.1 Fast medium | X AI | 10.0 | 6.2 | 3/3 | 5.65s |
| #1 | Gemini 3 Flash Preview medium | 10.0 | 10.0 | 3/3 | 5.61s | |
| #9 | GPT-5.4 medium | OpenAI | 10.0 | 8.0 | 3/3 | 5.02s |
| #11 | Claude Sonnet 4.6 medium | Anthropic | 7.0 | 7.7 | 2/3 | 4.95s |
| #25 | Claude Sonnet 4.6 none | Anthropic | 4.0 | 6.8 | 1/3 | 4.83s |
| #19 | GPT-5.3 Chat none | OpenAI | 7.3 | 7.3 | 2/3 | 4.72s |
| #3 | GPT-5.3-Codex medium | OpenAI | 10.0 | 8.4 | 3/3 | 4.69s |
| #48 | Qwen3 Coder Next none | Qwen | 2.3 | 4.0 | 0/3 | 4.39s |
| #15 | GPT-5.2 Chat none | OpenAI | 10.0 | 7.4 | 3/3 | 3.97s |
| #6 | Gemini 3 Pro Preview medium | 10.0 | 8.2 | 3/3 | 3.75s | |
| #45 | Trinity Large Preview none | Arcee AI | 10.0 | 4.2 | 0/3 | 3.59s |
| #5 | Gemini 3 Flash Preview low | 10.0 | 8.2 | 3/3 | 3.50s | |
| #31 | GLM 5 none | Z.ai | 4.0 | 6.0 | 1/3 | 3.39s |
| #29 | Qwen3.5 Plus 2026-02-15 none | Qwen | 4.0 | 6.2 | 1/3 | 2.74s |
| #12 | Gemini 3.1 Flash Lite Preview medium | 9.0 | 7.5 | 2/3 | 2.53s | |
| #17 | Gemini 3.1 Flash Lite Preview low | 7.0 | 7.3 | 2/3 | 2.18s | |
| #47 | GPT-4o-mini none | OpenAI | 4.0 | 4.0 | 1/3 | 1.83s |
| #42 | Qwen3.5-35B-A3B none | Qwen | 10.0 | 4.7 | 0/3 | 1.76s |
| #53 | Grok 4.1 Fast none | X AI | 1.3 | 2.9 | 0/3 | 1.73s |
| #37 | Qwen3.5-Flash none | Qwen | 2.3 | 5.2 | 0/3 | 1.62s |
| #20 | Gemini 3 Flash Preview none | 7.0 | 7.2 | 2/3 | 1.59s | |
| #44 | GPT-5.4 none | OpenAI | 10.0 | 4.5 | 0/3 | 1.41s |
| #54 | MiMo-V2-Flash none | Xiaomi | 10.0 | 2.9 | 0/3 | 1.36s |
| #36 | Mercury 2 medium | Inception | 7.3 | 5.3 | 2/3 | 1.30s |
| #22 | Gemini 3.1 Flash Lite Preview none | 6.0 | 7.1 | 1/3 | 1.16s | |
| #40 | Qwen3.5-122B-A10B none | Qwen | 4.0 | 5.0 | 1/3 | 927ms |
| #41 | Qwen3.5-27B none | Qwen | 4.0 | 4.9 | 1/3 | 796ms |
| #38 | Gemini 2.5 Flash none | 10.0 | 5.2 | 0/3 | 668ms | |
| #55 | LFM2-24B-A2B none | Liquid | 10.0 | 2.6 | 0/3 | 471ms |
| #51 | Mercury 2 none | Inception | 10.0 | 3.4 | 0/3 | 466ms |