Those you who have been following us in the past few months know that we are big fans of Kimi K2. It is a wonderful open source thinking model that can perform on par or better than closed models in humanity’s last exam, LiveCodeBench, and SWW-Multilingual Agentic coding.
๐ Hello, Kimi K2 Thinking!
The Open-Source Thinking Agent Model is here.๐น SOTA on HLE (44.9%) and BrowseComp (60.2%)
๐น Executes up to 200 โ 300 sequential tool calls without human interference
๐น Excels in reasoning, agentic search, and coding
๐น 256K context windowBuiltโฆ pic.twitter.com/lZCNBIgbV2
โ Kimi.ai (@Kimi_Moonshot) November 6, 2025
Kimi is great in reasoning and coding. It has a 256k context window. As Artificial Analysis explains on X, “Kimi K2 Thinking achieves 93% in ๐ยฒ-Bench Telecom, an agentic tool use benchmark where the model acts as a customer service agent.” It is #1 over GPT and MiniMax-M2 in agentic tool use.
[HT]
