Grok-4 and Kimi K2 Coming Soon to MyShell
We’re continually expanding ShellAgent with the latest high-performance models. Our engineers are already testing the newest LLMs for seamless integration.
• @Grok 4 currently tops the ARC-AGI benchmark with 15.9%, nearly doubling the next best score — a result that suggests it's one of the most capable AI models available today.
• @Kimi_Moonshot K2, one of the most powerful open-source LLMs available, outperforming AI giants like GPT-4.1 and Claude Opus 4 in benchmarks for coding, math, tool use, and agentic tasks. We believe K2’s robust agentic-task skills will shine inside ShellAgent.
Once their APIs are live and stable, we’ll wire them straight into ShellAgent.
Stay tuned, shellz!