Grok 4.3 vs Step 5 Preview
Wins 0 of 6 areas
—
Wins 4 of 6 areas
Coding · Agents · Reasoning · Long documents
Step 5 Preview is the stronger all-rounder.
Scores updated · 15 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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.
- AgentsCarrying out multi-step tasks on its own04Step 5 Preview4 of 4 tests
- CodingWriting and fixing software02Step 5 Preview2 of 2 tests
- ReasoningHard problems that need careful thinking02Step 5 Preview2 of 2 tests
- Long documentsFinding answers in very long texts01Step 5 Preview1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
- Images and chartsUnderstanding pictures, charts and video11Even1 each
Long documents rests on a single test.
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.
Step 5 Preview costs 1% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Step 5 Preview pulls ahead
- Real work tasks from 44 professionsGDPVal+32.4points ahead
- Finds the root cause of IT system incidentsAA IT-Bench SRE+22.9points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+21points ahead
Where Grok 4.3 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+17.2points ahead
- Harder college exam questions with imagesMMMU-Pro+1.7points ahead
Every test, side by side
All 15 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingStep 5 Preview
- LMArena · WebDevStep 5 Preview by 214 rating points13571571+214 rating
- SciCodeStep 5 Preview by 10.648.358.9+10.6
AgentsStep 5 Preview
- Terminal-Bench 4.0Step 5 Preview by 33.3033.3+33.3
- GDPValStep 5 Preview by 32.421.954.3+32.4
- AA IT-Bench SREStep 5 Preview by 22.932.755.6+22.9
- AA ApexAgentsStep 5 Preview by 211738+21
ReasoningStep 5 Preview
- CritPtStep 5 Preview by 12.9820.9+12.9
- Humanity's Last ExamStep 5 Preview by 9.337.246.5+9.3
FactsEven
- AA-Omniscience · Non-hallucinationGrok 4.3 by 17.274.257+17.2
- AA-Omniscience · AccuracyStep 5 Preview by 6.734.841.5+6.7
Images and chartsEven
- LMArena · VisionStep 5 Preview by 36 rating points12291265+36 rating
- MMMU-ProGrok 4.3 by 1.778.176.4+1.7
Other results2 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.
- AA IntelligenceStep 5 Preview by 18.824.943.7+18.8
- AA-OmniscienceGrok 4.3 by 1.61816.4+1.6
Questions people ask
Which is better, Grok 4.3 or Step 5 Preview?
Step 5 Preview wins four of the six areas where both have results: coding, agents, reasoning and long documents. Grok 4.3 wins none. They are level on facts and images and charts.
Which is better for coding?
Step 5 Preview. It wins 2 of the 2 coding tests both models report; Grok 4.3 wins none.
Which is cheaper?
Grok 4.3 costs $1.25 per million input tokens and $2.50 per million output tokens; Step 5 Preview costs $1.00 and $2.70. That makes Step 5 Preview about 1% cheaper for the same work.
How do you compare the two?
We use the 15 benchmark tests both models have published scores on. The verdict counts the 13 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 2 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.