Gemini 3 Pro vs gpt-oss-20b
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-20b is cheaper.
Scores updated · 28 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.
- CodingWriting and fixing software50Gemini 3 Pro5 of 5 tests
- ReasoningHard problems that need careful thinking30Gemini 3 Pro3 of 3 tests
- AgentsCarrying out multi-step tasks on its own20Gemini 3 Pro2 of 2 tests
- FactsGetting facts right instead of making them up20Gemini 3 Pro2 of 2 tests
- Long documentsFinding answers in very long texts10Gemini 3 Pro1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemini 3 Pro1 of 1 test
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.
gpt-oss-20b costs 98% 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
- Reasons across sets of long documentsAA-LCR+41.3points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+39.8points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+31.1points ahead
Where gpt-oss-20b pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 28 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3 Pro
- Terminal-Bench HardGemini 3 Pro by 31.141.710.6+31.1
- LiveCodeBench v6Gemini 3 Pro by 26.487.461+26.4
- SWE-bench VerifiedGemini 3 Pro by 23.876.252.4+23.8
- SWE-bench MultilingualGemini 3 Pro by 23.16541.9+23.1
- SciCodeGemini 3 Pro by 17.256.138.9+17.2
AgentsGemini 3 Pro
- GDPValGemini 3 Pro by 34.234.20+34.2
- BrowseCompGemini 3 Pro by 30.959.228.3+30.9
ReasoningGemini 3 Pro
- Humanity's Last ExamGemini 3 Pro by 28.739.711+28.7
- GPQA DiamondGemini 3 Pro by 2290.868.8+22
- CritPtGemini 3 Pro by 7.79.11.4+7.7
FactsGemini 3 Pro
- AA-Omniscience · AccuracyGemini 3 Pro by 39.855.816+39.8
- AA-Omniscience · Non-hallucinationGemini 3 Pro by 2.68.55.9+2.6
Following instructionsGemini 3 Pro
- IFBenchGemini 3 Pro by 5.370.465.1+5.3
Other results14 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 78.315.3-63+78.3
- τ²-BenchGemini 3 Pro by 50.39847.7+50.3
- τ²-Bench Telecom (AA run)Gemini 3 Pro by 26.987.160.2+26.9
- HarmBenchgpt-oss-20b by 26.272.598.7+26.2
- Artificial Analysis Coding IndexGemini 3 Pro by 25.846.520.7+25.8
- HMMT Feb. 2025Gemini 3 Pro by 20.897.576.7+20.8
- AA IntelligenceGemini 3 Pro by 19289+19
- MMLU-ProGemini 3 Pro by 13.790.176.4+13.7
- AIR-Bench 2024gpt-oss-20b by 12.873.286+12.8
- AIME 2025Gemini 3 Pro by 10.810089.2+10.8
- XSTestGemini 3 Pro by 10.197.387.2+10.1
- anthropic_red_teamgpt-oss-20b by 2.69799.7+2.6
- simple_safety_testsgpt-oss-20b by 2.597.5100+2.5
- bbqGemini 3 Pro by 1.798.496.7+1.7
Questions people ask
Which is better, Gemini 3 Pro or gpt-oss-20b?
Gemini 3 Pro wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. gpt-oss-20b wins none, but costs 98% less.
Which is better for coding?
Gemini 3 Pro. It wins 5 of the 5 coding tests both models report; gpt-oss-20b wins none.
Which is cheaper?
Gemini 3 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; gpt-oss-20b costs $0.07 and $0.18. That makes gpt-oss-20b about 98% cheaper for the same work.
How do you compare the two?
We use the 28 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.