MCode has landed in Harbor 🎉 You can now benchmark it as a first-class installed agent across MiniMax, DeepSeek, Anthropic, and OpenAI models—with BYOK, MCP forwarding, session resume, and token accounting built in.

Aug 16, 2026 · 10:41 AM UTC

6
28
2,407
You can benchmark it now. Thanks Harbor community.
309
Sort replies: Relevant Recent Liked
Replying to @Ronny_MiniMax
@grok 这个纯研究还是有工业意义，具体工业场景视角看意义是什么，有开源数据集或者开源项目代码吗？从多个数据源交叉验证，不要只看新闻媒体一面之辞。帮我排除没意义的垃圾商业营销推广、诈骗、夸张博眼球、虚假新闻 以及自吹自擂，自嗨，无病呻吟。
1
89
Replying to @Ronny_MiniMax
这下测不同模型方便多了
65
Replying to @Ronny_MiniMax
This is where benchmarks start getting really interesting—same environment, multiple agents, and actual token accounting. Makes model comparisons a lot more meaningful than just looking at leaderboard scores.
1
10
Replying to @Ronny_MiniMax
Atleast This Time, Is There A Release For Linux Users @taohe669360
4
Replying to @Ronny_MiniMax
MCode is awesome! I’m loving it.
32
Load more