Gemini 3.1 Pro vs Gemini 4 Argon
Wins 1 of 7 areas
Long documents
Wins 5 of 7 areas
Coding · Agents · Reasoning · Math · Following instructions
Gemini 4 Argon is the stronger all-rounder.Gemini 3.1 Pro is better at long documents.
Both rank among the ten best models we track in facts.
Scores updated · 35 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- ReasoningHard problems that need careful thinking04Gemini 4 Argon4 of 4 tests
- CodingWriting and fixing software02Gemini 4 Argon2 of 2 tests
- AgentsCarrying out multi-step tasks on its own02Gemini 4 Argon2 of 2 tests
- MathCompetition and research-level math02Gemini 4 Argon2 of 2 tests
- Following instructionsDoing exactly what it is asked01Gemini 4 Argon1 of 1 test
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 1 test
- FactsGetting facts right instead of making them up22Even2 each
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 4 Argon costs 14% 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 Gemini 4 Argon pulls ahead
- Complex command-line tasks across many fieldsTerminal-Bench 4.0+53.1points ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+51.3points ahead
- Solves new puzzle games it has never seenARC-AGI-3+45.9points ahead
Where Gemini 3.1 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+5points ahead
- Reasons across sets of long documentsAA-LCR+2.3points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 4 Argon
- LMArena · WebDevGemini 4 Argon by 232 rating points14471679+232 rating
- SciCodeGemini 4 Argon by 3.158.761.8+3.1
AgentsGemini 4 Argon
- Terminal-Bench 4.0Gemini 4 Argon by 53.1457.1+53.1
- GDPValGemini 4 Argon by 41.614.656.2+41.6
ReasoningGemini 4 Argon
- ARC-AGI-3Gemini 4 Argon by 45.90.446.3+45.9
- ARC-AGI-2Gemini 4 Argon by 14.277.191.3+14.2
- Humanity's Last ExamGemini 4 Argon by 10.14757.1+10.1
- CritPtGemini 4 Argon by 9.417.727.1+9.4
FactsEven
- AA-Omniscience · Non-hallucinationGemini 4 Argon by 35.849.184.9+35.8
- AA-Omniscience · AccuracyGemini 3.1 Pro by 554.949.9+5
- SimpleQA VerifiedGemini 4 Argon by 2.573.576+2.5
- Vectara HHEM hallucination ratelower is betterGemini 3.1 Pro by 1.710.412.1+1.7
MathGemini 4 Argon
- FrontierMath Tier 4Gemini 4 Argon by 51.326.878+51.3
- FrontierMath Tiers 1-3 (v2)Gemini 4 Argon by 26.359.686+26.3
Following instructionsGemini 4 Argon
- Multi-ChallengeGemini 4 Argon by 7.371.478.7+7.3
Other results19 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 1.1Gemini 4 Argon by 65.91277.9+65.9
- HiL-Bench (Tools-allowed)Gemini 4 Argon by 26.735.362+26.7
- LVBenchGemini 4 Argon by 25.566.291.7+25.5
- PostTrainBenchGemini 4 Argon by 23.721.645.3+23.7
- AA IntelligenceGemini 4 Argon by 22.929.752.6+22.9
- ERQAGemini 4 Argon by 16.270.887+16.2
- HealthBench HardGemini 4 Argon by 13.631.845.4+13.6
- HealthBench Pro, with Search groundingGemini 4 Argon by 12.76476.7+12.7
- HealthBench Pro, without Search groundingGemini 4 Argon by 11.657.268.8+11.6
- AA-OmniscienceGemini 4 Argon by 10.531.942.4+10.5
- HealthBench Pro, with Search grounding (length-adjusted)Gemini 4 Argon by 9.152.361.4+9.1
- vectara_avg_summary_lengthGemini 4 Argon by 70.5 rating points107.7178.2+70.5 rating
- CoT-Control (continuous controllability, average over all datasets and constraints)Gemini 4 Argon by 5.592.397.8+5.5
- CoT-Control (continuous controllability, averaged over GPQA, HLE, MMLU-Pro)Gemini 4 Argon by 5.592.397.8+5.5
- CoT-Control (strict controllability, average over all datasets and constraints)Gemini 4 Argon by 4.878.983.7+4.8
- CoT-Control (strict controllability, averaged over GPQA, HLE, MMLU-Pro)Gemini 4 Argon by 4.878.983.7+4.8
- vectara_factual_consistencyGemini 3.1 Pro by 1.789.687.9+1.7
- ARC-AGI-1tie9898.5tie
- vectara_answer_ratetie99.499.1tie
Questions people ask
Which is better, Gemini 3.1 Pro or Gemini 4 Argon?
Gemini 4 Argon wins five of the seven areas where both have results: coding, agents, reasoning, math and following instructions. Gemini 3.1 Pro wins long documents. They are level on facts.
Which is better for coding?
Gemini 4 Argon. It wins 2 of the 2 coding tests both models report; Gemini 3.1 Pro wins none.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Gemini 4 Argon costs $2.00 and $10.00. That makes Gemini 4 Argon about 14% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 16 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 19 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.