The Model Index
Models/xAI

Grok 4.20

xAIClosed weightsFlagship

Grok 4.20 is a closed AI model from xAI with an index score of 156 (GPT-4 = 100), rank 57 of 217.

156140–172Index score#57 of 217 models
$1.25/ $2.5Price per million tokensinput / output · via OpenRouter
2MContext windowtokens it reads at once
17 Feb 2026ReleasedReplaced by Grok 4.3

Where it sitson the frontier

801001201401601802002023202420252026GPT-4 at launch = 100Grok 4.20

Gemini 3 Pro reached this score 3 months earlier.

Test resultsby area · tick = best by any model

Reasoning & knowledgeevidence
GPQA Diamond (Epoch AI)
89%
Humanity's Last Exam
no result
Mathsevidence
FrontierMath (tiers 1-3)
45%
Codingevidence
SWE-bench Verified
no result
Terminal-Bench 4.0
no result
Agentsevidence
APEX-Agents
no result
Professional workevidence
Finance Agent (v2)
28%
Novel problemsevidence
ARC-AGI-2
65%
Long tasksno evidence yet
METR time horizon
no result

Third-party results only, at the model’s best standard setting. Hover a bar for who ran it.

What others sayrank on 2 leaderboards

The Model Indexthird-party tests, our method#57of 217 · 156
Epoch Capabilities Indexoverall capability from many benchmarks#59of 231 · 152.3

Best attop three among current models

Replaced by Grok 4.3; it no longer leads any category among current models.

Where to use itprices checked 18 Nov 2026

API providers 1 · per 1M tokens, in / out

Host prices from OpenRouter; going direct can differ.

Facts and sources

Released17 Feb 2026
Announcementdocs.x.ai ↗
Release entryGrok 4.20 (beta; GA/API 03-10)
Scored asGrok 4.20
Output speed–
Hugging Faceclosed weights

xAI releasesaround this one

This page as a chartDownload PNG