Gemini 3 Pro vs gpt-oss-120b
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
Gemini 3 Pro is the stronger all-rounder.gpt-oss-120b is cheaper.
Scores updated · 33 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.
- ReasoningHard problems that need careful thinking40Gemini 3 Pro4 of 4 tests
- CodingWriting and fixing software30Gemini 3 Pro3 of 3 tests
- FactsGetting facts right instead of making them up20Gemini 3 Pro2 of 4 tests · 2 ties
- AgentsCarrying out multi-step tasks on its own20Gemini 3 Pro2 of 2 tests
- Following instructionsDoing exactly what it is asked20Gemini 3 Pro2 of 2 tests
- Long documentsFinding answers in very long texts10Gemini 3 Pro1 of 1 test
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.
gpt-oss-120b costs 95% 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
- Short factual questions, answered correctlySimpleQA Verified+59points ahead
- Common-sense trick questionsSimpleBench+54.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+34points ahead
Where gpt-oss-120b pulls ahead
No clear win on a test scored out of 100.
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.
CodingGemini 3 Pro
- SWE-bench ProGemini 3 Pro by 27.143.316.2+27.1
- SciCodeGemini 3 Pro by 22.156.134+22.1
- Terminal-Bench HardGemini 3 Pro by 18.241.723.5+18.2
AgentsGemini 3 Pro
- GDPValGemini 3 Pro by 28.634.25.6+28.6
- AA ApexAgentsGemini 3 Pro by 15.318.43.1+15.3
ReasoningGemini 3 Pro
- SimpleBenchGemini 3 Pro by 54.376.422.1+54.3
- Humanity's Last ExamGemini 3 Pro by 20.139.719.6+20.1
- GPQA DiamondGemini 3 Pro by 12.690.878.2+12.6
- CritPtGemini 3 Pro by 89.11.1+8
FactsGemini 3 Pro
- SimpleQA VerifiedGemini 3 Pro by 5972.913.9+59
- AA-Omniscience · AccuracyGemini 3 Pro by 3455.821.8+34
- AA-Omniscience · Non-hallucinationtie8.59.2tie
- Vectara HHEM hallucination ratelower is bettertie13.614.2tie
Following instructionsGemini 3 Pro
- Multi-ChallengeGemini 3 Pro by 20.465.745.3+20.4
- IFBenchGemini 3 Pro by 1.470.469+1.4
Other results17 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 64.515.3-49.3+64.5
- AA Agentic IndexGemini 3 Pro by 45.8526.2+45.8
- HarmBenchgpt-oss-120b by 27.572.5100+27.5
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 21.387.165.8+21.3
- AA IntelligenceGemini 3 Pro by 16.42811.6+16.4
- Artificial Analysis Coding IndexGemini 3 Pro by 16.146.530.4+16.1
- AIR-Bench 2024gpt-oss-120b by 14.873.288+14.8
- AIME 2025Gemini 3 Pro by 1010090+10
- XSTestGemini 3 Pro by 4.697.392.7+4.6
- vectara_avg_summary_lengthgpt-oss-120b by 33.3 rating points101.9135.2+33.3 rating
- HMMT Nov. 2025Gemini 3 Pro by 3.393.390+3.3
- simple_safety_testsgpt-oss-120b by 2.597.5100+2.5
- anthropic_red_teamgpt-oss-120b by 2.49799.5+2.4
- HMMT Feb. 2025tie97.596.7tie
- vectara_factual_consistencytie86.485.8tie
- vectara_answer_ratetie99.499.9tie
- bbqtie98.498.5tie
Questions people ask
Which is better, Gemini 3 Pro or gpt-oss-120b?
Gemini 3 Pro wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. gpt-oss-120b wins none, but costs 95% less.
Which is better for coding?
Gemini 3 Pro. It wins 3 of the 3 coding tests both models report; gpt-oss-120b wins none.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; gpt-oss-120b costs $0.15 and $0.59. That makes gpt-oss-120b about 95% 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 16 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.