跳过主要内容
评测来源/Code Migration
参与排名编程

Code Migration

Vals AI · Code Migration 的公开模型评测结果。

官方评测
来源说明

该来源测量模型在「编程」领域的具体交付表现,原始成绩会和统一能力读数分开展示。

限制: 不同版本、推理设置和测试集可能导致分数变化,请结合数据时间阅读。

使用方式

参与排名的来源会影响能力分;参考、交叉参考和观察来源仅用于对比与核验。

本来源的模型成绩

Claude Sonnet 5.569.8378.72026-11-05
Gemini 4 Argon68.1777.42026-11-05
GPT-6 Astra67.7477.02026-11-05
Claude Opus 5.566.6576.22026-11-05
GPT-6.1 Sol65.1274.92026-11-05
Claude Opus 557.4768.82026-11-05
GPT-6 Sol57.268.62026-11-05
Claude Fable 555.0666.92026-11-05
Claude Fable 5.154.6166.52026-11-05
GPT-5.6 Sol52.9265.22026-11-05
GPT-5.6 Terra47.861.12026-11-05
Muse Spark 1.347.4160.82026-11-05
Claude Opus 4.847.2560.62026-11-05
DeepSeek V4.1 Flash45.6259.32026-11-05
GPT-5.545.1659.02026-11-05
Grok 4.744.8258.72026-11-05
Grok 4.644.5758.52026-11-05
Claude Haiku 5.544.5758.52026-11-05
GPT-5.6 Luna44.5558.52026-11-05
Claude Sonnet 544.3958.32026-11-05
GLM 5.344.2258.22026-11-05
DeepSeek V4.1 Flash 101744.2158.22026-11-05
Claude Opus 4.743.8857.92026-11-05
MiMo V2.6 Pro43.0157.22026-11-05
GPT-6 Luna42.5556.92026-11-05
DeepSeek V4 Pro 081341.5456.12026-11-05
MiMo V2.6 Flash40.9355.62026-10-17
MiMo V2.6 Flash40.9355.62026-10-18
MiMo V2.6 Flash40.9355.62026-11-05
Claude Sonnet 4.639.8954.72026-11-05
Step 5 Preview38.9854.02026-11-05
DeepSeek V4 Flash 073138.6353.72026-11-05
GLM-5.237.8753.12026-11-05
Grok 4.536.652.12026-11-05
Gemini 3.8 Flash36.5552.12026-11-05
GPT-5.434.9850.82026-11-05
Muse Spark 1.131.1147.72026-11-05
Gemini 3.6 Flash30.9347.62026-11-05
Mistral Large 430.5647.32026-10-31
Muse Spark 1.229.9546.82026-11-05
Kimi K2.627.7745.02026-11-05
GLM-5.125.7743.42026-11-05
Kimi K2.7 Code25.3943.12026-11-05
Qwen3.8 Max23.9642.02026-11-05
MiMo-V2.5-Pro21.5640.12026-11-05
GLM-5.3 Flash20.5239.22026-10-18
GLM-5.3 Flash20.5239.22026-10-17
GLM-5.3 Flash20.5239.22026-11-05
MiniMax-M319.9338.82026-11-05
Kimi K316.135.72026-11-05
MiMo-V2.514.2334.22026-11-05
Inkling Small13.6833.72026-11-05
Qwen3.7 Max13.0733.32026-11-05
GPT-5.4 mini12.9433.22026-11-05
Qwen3.7 Plus12.8633.12026-11-05
Inkling11.7932.22026-11-05
Qwen3.6 Plus11.131.72026-11-05
MiniMax-M2.78.6929.72026-11-05
Claude Haiku 4.57.5528.82026-11-05
Kimi K2.56.9628.42026-11-05
Grok 4.36.7928.22026-11-05
Gemini 3.5 Flash Lite6.1527.72026-11-05
Mistral Medium 3.55.1326.92026-10-11
Gemini 3.1 Flash Lite Preview4.6126.52026-11-05
Grok 4.20 (Reasoning)0.3123.02026-11-05
Ling-3.0-flash (free)022.82026-11-05

常见问题