Grok 3 Beta is the stronger all-rounder.
Scores updated · 11 tests both models report · How we compare
Where each one wins
Tests won in each of the two areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
Reasoning and images and charts rest on a single test each.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Grok 3 Beta pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+19.4points ahead
- College exam questions with charts, maps and diagramsMMMU+4.9points ahead
Where Claude 3.5 Sonnet pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 11 tests both models report. The winning score is in its model's colour; marks a score checked independently.
Other results9 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- HarmBenchClaude 3.5 Sonnet by 52.898.145.3+52.8
- AIR-Bench 2024Claude 3.5 Sonnet by 34.685.951.3+34.6
- LiveCodeBenchGrok 3 Beta by 24.232.857+24.2
- SimpleQAGrok 3 Beta by 15.228.443.6+15.2
- Aider-PolyglotGrok 3 Beta by 845.353.3+8
- anthropic_red_teamClaude 3.5 Sonnet by 4.399.895.5+4.3
- simple_safety_testsClaude 3.5 Sonnet by 3.210096.8+3.2
- MMLU-ProGrok 3 Beta by 2.377.679.9+2.3
- XSTesttie95.696.4tie
Questions people ask
Which is better, Claude 3.5 Sonnet or Grok 3 Beta?
Grok 3 Beta wins all two areas where both have results: reasoning and images and charts. Claude 3.5 Sonnet wins none.
Which is cheaper?
Claude 3.5 Sonnet costs $3.00 per million input tokens and $15.00 per million output tokens; Grok 3 Beta costs $3.00 and $15.00. For a million tokens read plus a million written, they cost about the same.
How do you compare the two?
We use the 11 benchmark tests both models have published scores on. The verdict counts the 2 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 9 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.