OpenClaw Rolls Out Support For Models Competing With Google Deepmind, OpenAI

AI platform OpenClaw has released a set of updates designed to make its tools faster and easier to use. The latest improvements include shell completion, which helps users type commands faster, and fixes for Telegram, improving message threading and how content is displayed.

The platform is also offering KIMI K2.5 and Kimi Coding for free, along with MiniMax, which can be accessed with just a single login. This allows developers to switch between AI models easily and run them directly on their own computers, giving them full control over their data.

“Kimi K2.5 + Kimi Coding: run your claw for free,” OpenClaw wrote in an X post. It also mentioned that another coding plan, MiniMax OAuth, was “just a login away.” 

OpenClaw also implemented a variety of community-suggested fixes across apps like LINE and BlueBubbles, as well as updates for routing, security, and login systems.

???? OpenClaw 2026.1.30

???? Shell completion
???? Kimi K2.5 + Kimi Coding: run your claw for free
???? MiniMax OAuth: one more model just a login away
???? Telegram got a glow-up — 6 fixes from threading to HTML rendering

Plus a bunch of community-contributed fixes across LINE,…

— OpenClaw???? (@openclaw) January 31, 2026

The strategy is “counterintuitive”, as per an X user. Most platforms lock users into one model. OpenClaw's strategy is to make model switching easier so users do not feel trapped. The more models are available for free use, the stickier the platform becomes.

OpenClaw just did something no major AI lab would do: made Kimi K2.5 and Kimi Coding free out of the box, and added MiniMax with a single OAuth login.

The strategy here is counterintuitive. Most platforms lock you into one model. OpenClaw is doing the opposite: making model… https://t.co/EFjMiTT50S

— Aakash Gupta (@aakashgupta) February 1, 2026

“The real competition in AI isn't between frontier labs anymore. It's between renting intelligence from cloud providers versus owning it on your hardware,” the post read.

The news comes amid Alibaba Group Holding and Moonshot AI unveiling their latest flagship AI models, which narrow the gap with US industry leaders Google DeepMind and OpenAI.

Moonshot AI announced its latest model, Kimi K2.5, on January 27, calling it the world's most powerful open-source model, the South China Morning Post reported. The Alibaba-backed startup introduced a feature called agent swarm in its new model, allowing developers to tackle complex coding tasks more easily by using up to 100 subagents to execute workflows in parallel.

Alibaba Cloud announced its biggest-ever model, Qwen3-Max-Thinking, on January 26, promising stronger agentic and tool-use capabilities.

While OpenClaw has gained attention for rolling out support to different models, it also poses security risks. The AI-powered agent often requires deep access equivalent to administrator privileges on the system it operates. In this context, if the agent is compromised or misunderstands instructions, damage can occur, as per Forbes.

AI-powered agents have also caused concern among researchers. A paper published in Science warned that AI agents could invade social media platforms in vast numbers and would form a new front in information warfare. Capable of mimicking human behaviour, these ‘AI swarms' could harass users, spread false narratives, and even undermine democracy.

The research was published in the backdrop of a new social media platform used by AI agents to share and discuss topics. Called Moltbook, the platform allows humans to observe only, and they cannot post anything.