Grok 4.20 (non-reasoning)

At a glance

What Grok 4.20 (non-reasoning) does

Vision, Tool use, Streaming, Structured output, Prompt caching

Best for

Latency-sensitive 1M-context work that does not need reasoning

Works well with

Conflicts & caveats

• Trae's built-in model menu may not include Grok 4.20 (non-reasoning) — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
• Google Antigravity is built around Gemini 3 Pro; Claude and recent GPT-5 models also work, but Grok 4.20 (non-reasoning) is not in the supported model list.
• Kiro's spec mode is tuned for Claude models — Grok 4.20 (non-reasoning) may work but spec-driven flows are best with Claude.
• Cursor's built-in model menu may not include Grok 4.20 (non-reasoning) — verify it's available in your Cursor model picker.
• OpenAI Codex (VS Code) is tied to OpenAI's hosted GPT-5 and Codex models — Grok 4.20 (non-reasoning) is not selectable inside this integration.
• Gemini CLI is locked to Google Gemini models — your LLM is Grok 4.20 (non-reasoning).
• Claude Code only works with Claude models — your LLM is Grok 4.20 (non-reasoning).
• Devin uses Cognition's proprietary model internally — your selected Grok 4.20 (non-reasoning) will be ignored.
• OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected Grok 4.20 (non-reasoning) will be ignored.
• Jules runs Gemini models internally — your selected Grok 4.20 (non-reasoning) will be ignored.
• The Claude Code SDK is tuned for Claude models — Grok 4.20 (non-reasoning) may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
• Lovable uses a fixed internal model — your selected Grok 4.20 (non-reasoning) will be ignored.
• Replit Agent 3 uses its own model selection internally — your selected Grok 4.20 (non-reasoning) will be ignored.