Gemini 3 Pro vs GPT-5.4
Wins 1 of 8 areas
Facts
Wins 6 of 8 areas
Coding · Agents · Reasoning · Math · Long documents · Following instructions
GPT-5.4 is the stronger all-rounder.Gemini 3 Pro is cheaper and better at facts.
Scores updated · 47 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.
- CodingWriting and fixing software04GPT-5.44 of 5 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own04GPT-5.44 of 4 tests
- ReasoningHard problems that need careful thinking04GPT-5.44 of 4 tests
- MathCompetition and research-level math03GPT-5.43 of 3 tests
- Long documentsFinding answers in very long texts01GPT-5.41 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5.41 of 1 test
- FactsGetting facts right instead of making them up21Gemini 3 Pro2 of 4 tests · 1 tie
- Images and chartsUnderstanding pictures, charts and video11Even1 each · 1 tie
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.
Gemini 3 Pro costs 20% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where GPT-5.4 pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+42.2points ahead
- Finds hard-to-locate facts by browsing the webBrowseComp+23.5points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+15.9points ahead
Where Gemini 3 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+27.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+4.9points ahead
- Harder college exam questions with imagesMMMU-Pro+1.8points ahead
Every test, side by side
All 47 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGPT-5.4
- Terminal-Bench HardGPT-5.4 by 15.941.757.6+15.9
- SWE-bench ProGPT-5.4 by 14.443.357.7+14.4
- SWE-bench MultilingualGPT-5.4 by 6.76571.7+6.7
- LMArena · WebDevGPT-5.4 by 25 rating points14401465+25 rating
- SciCodetie56.156.6tie
AgentsGPT-5.4
- BrowseCompGPT-5.4 by 23.559.282.7+23.5
- AA ApexAgentsGPT-5.4 by 14.918.433.3+14.9
- MCP AtlasGPT-5.4 by 13.154.167.2+13.1
- GDPValGPT-5.4 by 3.234.237.4+3.2
ReasoningGPT-5.4
- ARC-AGI-2GPT-5.4 by 42.231.173.3+42.2
- CritPtGPT-5.4 by 14.39.123.4+14.3
- Humanity's Last ExamGPT-5.4 by 439.743.7+4
- GPQA DiamondGPT-5.4 by 1.290.892+1.2
FactsGemini 3 Pro
- SimpleQA VerifiedGemini 3 Pro by 27.872.945.1+27.8
- Vectara HHEM hallucination ratelower is betterGPT-5.4 by 6.613.67+6.6
- AA-Omniscience · AccuracyGemini 3 Pro by 4.955.850.9+4.9
- AA-Omniscience · Non-hallucinationtie8.58.3tie
Images and chartsEven
- MMMU-ProGemini 3 Pro by 1.880.278.4+1.8
- CharXiv (RQ)GPT-5.4 by 1.481.482.8+1.4
- LMArena · Visiontie13051302tie
MathGPT-5.4
- HMMT Feb. 2026GPT-5.4 by 11.386.497.7+11.3
- IMOAnswerBenchGPT-5.4 by 8.383.191.4+8.3
- AIME 2026GPT-5.4 by 7.591.799.2+7.5
Other results22 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.
- GDPval-AA (Elo)GPT-5.4 by 479 rating points11951674+479 rating
- Artificial Analysis Coding IndexGPT-5.4 by 24.646.571.1+24.6
- browsecomp_with_context_managerGPT-5.4 by 23.559.282.7+23.5
- Terminal-Bench 2.0GPT-5.4 by 20.954.275.1+20.9
- ARC-AGI-1GPT-5.4 by 18.77593.7+18.7
- ToolathlonGPT-5.4 by 18.236.454.6+18.2
- DeepSearchQA (F1)GPT-5.4 by 15.463.278.6+15.4
- AA IntelligenceGPT-5.4 by 112839+11
- AA-OmniscienceGemini 3 Pro by 9.515.35.8+9.5
- frontiermath_tier_4_v1GPT-5.4 by 8.318.827.1+8.3
- matharena_visual_math_overallGPT-5.4 by 8.384.292.5+8.3
- AA Agentic IndexGemini 3 Pro by 7.85244.2+7.8
- LiveBenchGPT-5.4 by 6.973.480.3+6.9
- vectara_factual_consistencyGPT-5.4 by 6.686.493+6.6
- HLE (with tools)GPT-5.4 by 6.245.852+6.2
- MATH-VisionGPT-5.4 by 5.986.192+5.9
- MMLU-ProGemini 3 Pro by 2.690.187.5+2.6
- HMMT Nov. 2025GPT-5.4 by 2.593.395.8+2.5
- vectara_avg_summary_lengthGemini 3 Pro by 20.2 rating points101.981.7+20.2 rating
- OmniDocBench 1.5tie88.589.1tie
- vectara_answer_ratetie99.499.9tie
- τ²-Bench Telecom (AA run)tie87.187.1tie
Questions people ask
Which is better, Gemini 3 Pro or GPT-5.4?
GPT-5.4 wins six of the eight areas we test: coding, agents, reasoning, math, long documents and following instructions. Gemini 3 Pro wins facts, and costs 20% less. They are level on images and charts.
Which is better for coding?
GPT-5.4. It wins 4 of the 5 coding tests both models report; Gemini 3 Pro wins none, and 1 is a tie.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; GPT-5.4 costs $2.50 and $15.00. That makes Gemini 3 Pro about 20% cheaper for the same work.
How do you compare the two?
We use the 47 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 22 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.