Grok 4.20 Non-Reasoning is the speed-optimized production variant of xAI's March 2026 Grok 4.20 generation, returning direct answers without extended reasoning. It keeps a 2 million token context window and real-time integration with X data, enabling applications that need current information and low latency. The model supports native tool use, structured outputs, and image input across multimodal requests. Without the reasoning overhead of the chain-of-thought variants, the non-reasoning mode prioritizes response speed, reported at around 232 tokens per second output and roughly half-a-second time-to-first-token, making it a fit for interactive systems, support, and content generation. Released in March 2026, Grok 4.20 Non-Reasoning is xAI's high-throughput path for the 4.20 generation when you need capability, long context, and real-time grounding without the latency of reasoning steps.
Grok 4.20 Non-Reasoning runs on 3 different routes through the Opper gateway. Compare residency, ZDR, and training posture at a glance โ full data-handling detail per route below.
| Provider | Region | Zero data retention | Training | Input | Output |
|---|---|---|---|---|---|
| Google Cloud | Multi | Enterprise | No | $2.00 | $6.00 |
| xAI | US | Enterprise | No | $2.00 | $6.00 |
| xAI | EU | Enterprise | No | $2.00 | $6.00 |
Training posture across routes: No training on prompts by default.
Each route hosting Grok 4.20 Non-Reasoning has its own privacy posture, residency, and GDPR terms. Postures are maintained by Opper with a last-verification timestamp.
Zero data retention is available via Opper Enterprise contract. No training on customer data. US; SCCs; DPA available.
Zero data retention is available via Opper Enterprise contract. No training on customer data. US; SCCs; DPA available.
Zero data retention is available via Opper Enterprise contract. No training on customer data. EU; DPA available.
Independent benchmark scores โ composite indices for reasoning, coding, and math, plus individual eval scores where available.
Call Grok 4.20 Non-Reasoning through the Opper gateway with one API key. Let your coding agent set it up, or call it directly โ Opper is drop-in compatible with the OpenAI, Anthropic, and Google AI SDKs.
Copy this and paste it into your coding agent โ Claude Code, Cursor, Codex, and more โ and it'll wire up Opper for you.
import OpenAI from "openai";const client = new OpenAI({apiKey: process.env.OPPER_API_KEY,baseURL: "https://api.opper.ai/v3/compat",});const completion = await client.chat.completions.create({model: "xai/grok-4.20-non-reasoning",messages: [{ role: "user", content: "Hello" }],});console.log(completion.choices[0].message.content);
Side-by-side on privacy, EU hosting, pricing, and benchmarks.
