Original Summary

看左边这个Harvey’s Legal Agent Benchmark 模型名字 成绩 开销 用时 Muse Spark 1.2 25.42% $2.09 24m16s Muse Spark 1.3 Max 23.75% $2.26 14m43s Muse Spark 1.3 22.92% $2.76 17m04s Muse Spark 1.1 20.00% $0.80 12m21s Gemini 4 Argon 19.58% $10.55 43m55s GLM Legal 19.17% $0.15 1h12m Mistral Large 4 15.83% $4.18 39m25s Grok 4.6 15.83% $4.01 45m05s Grok 4.5 12.92% $2.02 10m15s Kimi K3 12.92% $3.83 13m58s Grok 4.7 12.50% $11.13 42m05s MiMo V2.6 Flash 11.25% $0.09 17m12s Qwen 3.8 27B 11.25% $4.03 18m39s Claude Fable 5 11.25% $19.23 26m53s MiMo V2.6 Pro 10.83% $0.22 25m53s Qwen 3.8 Max 10.42% $2.37 49m12s Gemini 3.8 Flash 10.00% $3.66 29m21s Claude Opus 4.8 9.58% $10.22 24m08s Step 5 Preview 9.17% $1.00 29m02s Hy4 Preview 9.17% $0.98 29m25s Ember-1 9.17% $3.33 44m26s Gemini 3.7 Flash 8.75% $2.56 7m32s DeepSeek V4 Flash 0731 8.33% $0.06 16m54s GLM 5.3 8.33% $4.30 42m38s DeepSeek V4 Pro 0813 7.50% $0.17 21m53s GLM 5.2 7.08% $2.06 20m51s DeepSeek V4.1 Flash 6.67% $0.17 9m39s Claude Opus 4.7 6.67% $10.81 24m44s GLM 5.3 Flash 6.67% $0.57 52m17s Claude Opus 5 6.67% $23.67 55m38s Claude Fable 5.1 6.67% $46.21 1h52m GPT-6 Astra 5.42% $26.16 24m49s GPT-6.1 Sol 5.42% $3.93 43m14s Claude Sonnet 4.6 5.00% $3.04 18m43s Claude Sonnet 5 5.00% $8.95 38m52s MiniMax-M3 4.17% $1.46 22m31s DeepSeek V4 3.75% $0.74 6m42s GPT 5.5 3.75% $4.60 12m14s Claude Opus 5.5 3.75% $21.38 52m55s Gemini 3.6 Flash 3.33% $1.90 16m24s GPT-6 Luna 2.92% $0.30 17m32s Claude Sonnet 5.5 2.92% $16.61 57m44s Gemini 3.5 Flash 2.50% $2.12 7m58s GPT-5.6 Sol 2.50% $10.37 22m50s MiMo V2.5 Pro 2.08% $0.11 9m04s Inkling 2.08% $1.25 18m24s MiMo V2.5 1.67% $0.04 6m59s GPT-6 Sol 1.67% $3.36 13m40s Kimi K2.6 1.67% $0.82 23m25s Inkling Small 1.67% $0.17 31m11s Qwen 3.7 Max 1.67% $1.58 32m03s Ling 3.0 Flash 1.25% $0.05 2m41s GPT-5.6 Luna 1.25% $0.38 10m56s Qwen 3.6 Plus 1.25% $3.94 14m48s Claude Haiku 5.5 1.25% $3.39 35m14s Claude Haiku 4.5 (Thinking) 0.83% $0.42 6m01s GPT-5.6 Terra 0.83% $3.20 17m05s Grok 4.3 0.42% $0.39 2m49s Nemotron 3 Ultra 0.42% N/A 34m09s Mistral Medium 3.5 0.42% $6.07 1h02m Gemini 3.1 Flash Lite Preview 0.00% $0.08 60.55s Mercury 2.5 0.00% $0.19 2m12s Gemini 3 Flash (12/25) 0.00% $0.26 3m23s Gemini 3.5 Flash Lite 0.00% $0.33 3m53s Ling 3.0 Flash Fin 0.00% $0.05 5m58s GPT 5.4 Nano 0.00% $0.18 6m03s MiniMax-M2.7 0.00% $0.49 6m14s Gemini 3.1 Pro Preview (02/26) 0.00% $1.15 7m17s Laguna M.1 0.00% N/A 7m45s GLM 5.1 0.00% $0.56 8m18s Grok 4.20 (Reasoning) 0.00% $0.99 8m23s Qwen 3.7 Plus 0.00% $0.23 8m49s GPT 5.4 Mini 0.00% $1.00 11m21s GPT 5.4 (xhigh) 0.00% $3.60 14m36s Laguna XS.2 0.00% N/A 19m54s Kimi K2.5 0.00% $0.35 25m08s Nemotron 3.5 Lightning 0.00% $0.03 28m56s 来源： Vals AI · Harvey’s Legal Agent Benchmark ，2026-10-12。 2 个帖子 - 1 位参与者 阅读完整话题

  • 情报分类：技术学习与提效
  • 分类依据：内容涉及技术、AI、软件工具或工程实践
  • 信息来源：服务器 / LINUX DO - 最新话题
  • 发布时间：2026/10/13 06:46:05