Gemini 3.1 Pro vs Grok 4.6
Wins 4 of 8 areas
Facts · Images and charts · Long documents · Following instructions
Wins 3 of 8 areas
Coding · Agents · Math
Gemini 3.1 Pro wins more areas, narrowly.Grok 4.6 is cheaper and better at agents.
Scores updated · 33 tests both models report · How we compare
Where each one wins
Tests won in each of the eight areas we test. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- FactsGetting facts right instead of making them up21Gemini 3.1 Pro2 of 3 tests
- Images and chartsUnderstanding pictures, charts and video10Gemini 3.1 Pro1 of 1 test
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemini 3.1 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own04Grok 4.64 of 4 tests
- MathCompetition and research-level math03Grok 4.63 of 3 tests
- CodingWriting and fixing software13Grok 4.63 of 5 tests · 1 tie
- ReasoningHard problems that need careful thinking33Even3 each · 1 tie
Images and charts, long documents and following instructions 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.
Grok 4.6 costs 43% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.1 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+24.6points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+11.9points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+10points ahead
Where Grok 4.6 pulls ahead
- Real work tasks from 44 professionsGDPVal+42.6points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+26.9points ahead
- Long professional tasks in banking, consulting and lawAA ApexAgents+25.5points ahead
Every test, side by side
All 33 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGrok 4.6
- LMArena · WebDevGrok 4.6 by 170 rating points14471617+170 rating
- Terminal-Bench 2.1Grok 4.6 by 14.273.888+14.2
- LiveBench · Agentic CodingGrok 4.6 by 12.944.157+12.9
- SciCodeGemini 3.1 Pro by 5.758.753+5.7
- LiveBench · Codingtie76.576.8tie
AgentsGrok 4.6
- GDPValGrok 4.6 by 42.614.757.3+42.6
- AA ApexAgentsGrok 4.6 by 25.53257.5+25.5
- τ-Bench V3 · BankingGrok 4.6 by 21.921.443.3+21.9
- Terminal-Bench 4.0Grok 4.6 by 13.2417.2+13.2
ReasoningEven
- ARC-AGI-2Gemini 3.1 Pro by 1077.167.1+10
- LiveBench · ReasoningGrok 4.6 by 6.58490.5+6.5
- SimpleBenchGemini 3.1 Pro by 3.779.675.9+3.7
- Humanity's Last ExamGemini 3.1 Pro by 2.94744.1+2.9
- CritPtGrok 4.6 by 217.719.7+2
- ARC-AGI-3Grok 4.6 by 1.70.42.1+1.7
- GPQA Diamondtie94.193.5tie
FactsGemini 3.1 Pro
- AA-Omniscience · Non-hallucinationGrok 4.6 by 26.949.176+26.9
- SimpleQA VerifiedGemini 3.1 Pro by 24.673.548.9+24.6
- AA-Omniscience · AccuracyGemini 3.1 Pro by 11.954.943+11.9
Images and chartsGemini 3.1 Pro
- LMArena · VisionGemini 3.1 Pro by 31 rating points12961265+31 rating
MathGrok 4.6
- FrontierMath Tiers 1-3 (v2)Grok 4.6 by 6.359.666+6.3
- FrontierMath Tier 4Grok 4.6 by 4.926.831.7+4.9
- LiveBench · MathematicsGrok 4.6 by 1.69192.6+1.6
Following instructionsGemini 3.1 Pro
- LiveBench · Instruction FollowingGemini 3.1 Pro by 7.279.171.9+7.2
Other results8 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.
- DeepSWE 1.1Grok 4.6 by 53.91265.9+53.9
- AA Agentic IndexGrok 4.6 by 42.410.352.7+42.4
- AA IntelligenceGrok 4.6 by 14.529.744.2+14.5
- ARC-AGI-1Gemini 3.1 Pro by 119887+11
- Artificial Analysis Coding IndexGrok 4.6 by 7.168.875.9+7.1
- livebench_data_analysisGemini 3.1 Pro by 4.678.573.9+4.6
- AA-OmniscienceGemini 3.1 Pro by 2.631.929.3+2.6
- livebench_languageGemini 3.1 Pro by 1.785.483.7+1.7
Questions people ask
Which is better, Gemini 3.1 Pro or Grok 4.6?
Gemini 3.1 Pro wins four of the eight areas we test: facts, images and charts, long documents and following instructions. Grok 4.6 wins coding, agents and math, and costs 43% less. They are level on reasoning.
Which is better for coding?
Grok 4.6. It wins 3 of the 5 coding tests both models report; Gemini 3.1 Pro wins 1, and 1 is a tie.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Grok 4.6 costs $2.00 and $6.00. That makes Grok 4.6 about 43% cheaper for the same work.
How do you compare the two?
We use the 33 benchmark tests both models have published scores on. The verdict counts the 25 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 8 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.