modelgrep

Best xAI Reasoning Models

Quick answer · Updated June 2026

The best xAI model for reasoning is Grok 4.3 — 53.2 intelligence with extended step-by-step thinking. Grok 4.20 (29.7) and Grok Build 0.1 (—) round out the top three.

53.2Intelligence
155 t/sSpeed
$1.25Input /M
1MContext

Large language models with extended reasoning / thinking, ranked by intelligence. The best models for complex multi-step reasoning, math and analysis.

  1. 1X
    grok-4.3
    ReasoningToolsJSON+153.2 intel · $1.25/M · 155 t/s
    53.2
    Intelligence
  2. 2X
    grok-4.20
    ReasoningToolsJSON+129.7 intel · $1.25/M · 76 t/s
    29.7
    Intelligence
  3. 3X
    grok-build-0.1
    ReasoningToolsJSON+1$1.00/M · 120 t/s · 769ms ttft
    Intelligence
  4. 4X
    grok-4.20-multi-agent
    ReasoningJSONVision$1.25/M · 325 t/s · 11.5s ttft
    Intelligence

Frequently asked

What is the best xAI model for reasoning?

The best xAI model for reasoning is Grok 4.3 — 53.2 intelligence with extended step-by-step thinking. Grok 4.20 (29.7) and Grok Build 0.1 (—) round out the top three.

What's a good alternative to Grok 4.3?

Grok 4.20 (29.7) is the closest alternative on this metric, followed by Grok Build 0.1 (—). See the full ranking above for the tradeoffs.

How many xAI models are there?

modelgrep tracks 4 xAI models with live benchmarks, speed, latency and per-provider pricing, led on intelligence by Grok 4.3. 4 of them qualify for this ranking.

More xAI rankings

All rankings