Gemini 3 Pro vs Gemma 4 26B A4B
Wins 6 of 8 areas
Coding · Agents · Reasoning · Images and charts · Math · Long documents
Wins 2 of 8 areas
Facts · Following instructions
Gemini 3 Pro is the stronger all-rounder.Gemma 4 26B A4B is cheaper and better at facts.
Scores updated · 45 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 software70Gemini 3 Pro7 of 7 tests
- Images and chartsUnderstanding pictures, charts and video40Gemini 3 Pro4 of 4 tests
- AgentsCarrying out multi-step tasks on its own30Gemini 3 Pro3 of 3 tests
- ReasoningHard problems that need careful thinking30Gemini 3 Pro3 of 3 tests
- MathCompetition and research-level math30Gemini 3 Pro3 of 3 tests
- Long documentsFinding answers in very long texts20Gemini 3 Pro2 of 2 tests
- FactsGetting facts right instead of making them up12Gemma 4 26B A4B2 of 3 tests
- Following instructionsDoing exactly what it is asked01Gemma 4 26B A4B1 of 1 test
Following instructions 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.
Gemma 4 26B A4B costs 96% 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 3 Pro pulls ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+58.8points ahead
- Fixes real GitHub issues in many programming languagesSWE-bench Multilingual+47.7points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+36.7points ahead
Where Gemma 4 26B A4B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+5.1points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+2points ahead
Every test, side by side
All 45 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3 Pro
- SWE-bench VerifiedGemini 3 Pro by 58.876.217.4+58.8
- SWE-bench MultilingualGemini 3 Pro by 47.76517.3+47.7
- SWE-bench ProGemini 3 Pro by 29.543.313.8+29.5
- Terminal-Bench HardGemini 3 Pro by 28.141.713.6+28.1
- SciCodeGemini 3 Pro by 16.156.140+16.1
- LiveCodeBench v6Gemini 3 Pro by 10.387.477.1+10.3
- LMArena · WebDevGemini 3 Pro by 81 rating points14401359+81 rating
AgentsGemini 3 Pro
- BrowseCompGemini 3 Pro by 32.959.226.3+32.9
- GDPValGemini 3 Pro by 30.834.23.4+30.8
- MCP AtlasGemini 3 Pro by 4.154.150+4.1
ReasoningGemini 3 Pro
- Humanity's Last ExamGemini 3 Pro by 20.439.719.3+20.4
- GPQA DiamondGemini 3 Pro by 11.690.879.2+11.6
- CritPtGemini 3 Pro by 9.19.10+9.1
FactsGemma 4 26B A4B
- AA-Omniscience · AccuracyGemini 3 Pro by 36.755.819.1+36.7
- Vectara HHEM hallucination ratelower is betterGemma 4 26B A4B by 8.413.65.2+8.4
- AA-Omniscience · Non-hallucinationGemma 4 26B A4B by 5.18.513.6+5.1
Images and chartsGemini 3 Pro
- CharXiv (RQ)Gemini 3 Pro by 12.481.469+12.4
- MMMU-ProGemini 3 Pro by 1180.269.2+11
- MathVistaGemini 3 Pro by 10.489.879.4+10.4
- LMArena · VisionGemini 3 Pro by 45 rating points13051260+45 rating
MathGemini 3 Pro
- IMOAnswerBenchGemini 3 Pro by 8.883.174.3+8.8
- HMMT Feb. 2026Gemini 3 Pro by 7.486.479+7.4
- AIME 2026Gemini 3 Pro by 3.491.788.3+3.4
Long documentsGemini 3 Pro
- MRCR v2 8 needle 128k (average)Gemini 3 Pro by 32.97744.1+32.9
- AA-LCRGemini 3 Pro by 10.37665.7+10.3
Following instructionsGemma 4 26B A4B
- IFBenchGemma 4 26B A4B by 270.472.4+2
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.
- AA-OmniscienceGemini 3 Pro by 66.115.3-50.8+66.1
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 43.587.143.6+43.5
- AA Agentic IndexGemini 3 Pro by 415211+41
- τ²-BenchGemini 3 Pro by 29.89868.2+29.8
- HLE (with tools)Gemini 3 Pro by 28.645.817.2+28.6
- Terminal-Bench 2.0Gemini 3 Pro by 2054.234.2+20
- SimpleVQAGemini 3 Pro by 17.569.752.2+17.5
- OmniDocBench 1.5Gemini 3 Pro by 14.188.574.4+14.1
- AA IntelligenceGemini 3 Pro by 11.32816.7+11.3
- vectara_factual_consistencyGemma 4 26B A4B by 8.486.494.8+8.4
- MMLU-ProGemini 3 Pro by 7.590.182.6+7.5
- Artificial Analysis Coding IndexGemini 3 Pro by 7.246.539.3+7.2
- HMMT Feb. 2025Gemini 3 Pro by 5.897.591.7+5.8
- HMMT Nov. 2025Gemini 3 Pro by 5.893.387.5+5.8
- MMMLUGemini 3 Pro by 5.591.886.3+5.5
- MATH-VisionGemini 3 Pro by 3.786.182.4+3.7
- vectara_avg_summary_lengthGemini 3 Pro by 34.8 rating points101.967.1+34.8 rating
- vectara_answer_ratetie99.499.8tie
- t2-benchtie85.485.5tie
Questions people ask
Which is better, Gemini 3 Pro or Gemma 4 26B A4B?
Gemini 3 Pro wins six of the eight areas we test: coding, agents, reasoning, images and charts, math and long documents. Gemma 4 26B A4B wins facts and following instructions, and costs 96% less.
Which is better for coding?
Gemini 3 Pro. It wins 7 of the 7 coding tests both models report; Gemma 4 26B A4B wins none.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Gemma 4 26B A4B costs $0.13 and $0.40. That makes Gemma 4 26B A4B about 96% cheaper for the same work.
How do you compare the two?
We use the 45 benchmark tests both models have published scores on. The verdict counts the 26 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.