Gemini 3.1 Pro vs Qwen3.6 Plus
Wins 7 of 8 areas
Coding · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 1 of 8 areas
Agents
Gemini 3.1 Pro is the stronger all-rounder.Qwen3.6 Plus is cheaper and better at agents.
Scores updated · 65 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 software73Gemini 3.1 Pro7 of 10 tests
- ReasoningHard problems that need careful thinking40Gemini 3.1 Pro4 of 4 tests
- MathCompetition and research-level math41Gemini 3.1 Pro4 of 5 tests
- Images and chartsUnderstanding pictures, charts and video30Gemini 3.1 Pro3 of 3 tests
- Following instructionsDoing exactly what it is asked20Gemini 3.1 Pro2 of 2 tests
- FactsGetting facts right instead of making them up21Gemini 3.1 Pro2 of 3 tests
- Long documentsFinding answers in very long texts10Gemini 3.1 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own12Qwen3.6 Plus2 of 4 tests · 1 tie
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.
Qwen3.6 Plus costs 75% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where Gemini 3.1 Pro pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+29.4points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+28.5points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+21.4points ahead
Where Qwen3.6 Plus pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.3points ahead
- Real work tasks from 44 professionsGDPVal+10points ahead
- Multi-step tasks using many real software toolsMCP Atlas+4.9points ahead
Every test, side by side
All 65 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.1 Pro
- SciCodeGemini 3.1 Pro by 1858.740.7+18
- Terminal-Bench 2.1Gemini 3.1 Pro by 12.473.861.4+12.4
- Terminal-Bench HardGemini 3.1 Pro by 9.953.843.9+9.9
- LiveCodeBench v6Gemini 3.1 Pro by 4.691.787.1+4.6
- SWE-bench MultilingualGemini 3.1 Pro by 3.176.973.8+3.1
- LiveBench · Agentic CodingGemini 3.1 Pro by 2.744.141.4+2.7
- SWE-bench ProQwen3.6 Plus by 2.454.256.6+2.4
- SWE-bench VerifiedGemini 3.1 Pro by 1.880.678.8+1.8
- LiveBench · CodingQwen3.6 Plus by 1.776.578.2+1.7
- LMArena · WebDevQwen3.6 Plus by 14 rating points14471461+14 rating
AgentsQwen3.6 Plus
- OSWorld-VerifiedGemini 3.1 Pro by 13.776.262.5+13.7
- GDPValQwen3.6 Plus by 1014.724.7+10
- MCP AtlasQwen3.6 Plus by 4.969.274.1+4.9
- τ-Bench V3 · Bankingtie21.420.8tie
ReasoningGemini 3.1 Pro
- Humanity's Last ExamGemini 3.1 Pro by 19.24727.8+19.2
- CritPtGemini 3.1 Pro by 14.817.72.9+14.8
- LiveBench · ReasoningGemini 3.1 Pro by 8.28475.8+8.2
- GPQA DiamondGemini 3.1 Pro by 5.994.188.2+5.9
FactsGemini 3.1 Pro
- SimpleQA VerifiedGemini 3.1 Pro by 29.473.544.1+29.4
- AA-Omniscience · AccuracyGemini 3.1 Pro by 28.554.926.4+28.5
- AA-Omniscience · Non-hallucinationQwen3.6 Plus by 16.349.165.4+16.3
Images and chartsGemini 3.1 Pro
- MMMU-ProGemini 3.1 Pro by 4.482.478+4.4
- Video-MMEGemini 3.1 Pro by 2.586.784.2+2.5
- CharXiv (RQ)Gemini 3.1 Pro by 1.883.381.5+1.8
MathGemini 3.1 Pro
- FrontierMath Tiers 1-3 (v2)Gemini 3.1 Pro by 21.459.638.3+21.4
- LiveBench · MathematicsGemini 3.1 Pro by 7.39183.7+7.3
- HMMT Feb. 2026Gemini 3.1 Pro by 6.994.787.8+6.9
- AIME 2026Gemini 3.1 Pro by 398.395.3+3
- IMOAnswerBenchQwen3.6 Plus by 2.88183.8+2.8
Following instructionsGemini 3.1 Pro
- LiveBench · Instruction FollowingGemini 3.1 Pro by 20.879.158.3+20.8
- IFBenchGemini 3.1 Pro by 1.977.175.2+1.9
Other results33 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 3.1 Pro by 3131.90.9+31
- AA Agentic IndexQwen3.6 Plus by 18.710.329+18.7
- DynaMathQwen3.6 Plus by 15.972.188+15.9
- Artificial Analysis Coding IndexGemini 3.1 Pro by 14.368.854.5+14.3
- livebench_languageGemini 3.1 Pro by 10.485.475+10.4
- LiveBenchGemini 3.1 Pro by 979.970.8+9
- Tool-DecathlonGemini 3.1 Pro by 948.839.8+9
- ToolathlonGemini 3.1 Pro by 948.839.8+9
- livebench_data_analysisGemini 3.1 Pro by 8.678.569.9+8.6
- frontiermath_tier_4_v1Gemini 3.1 Pro by 8.416.78.3+8.4
- AgentWorldBench - SearchGemini 3.1 Pro by 8.330.221.9+8.3
- MCPMarkGemini 3.1 Pro by 7.755.948.2+7.7
- Terminal-Bench 2.0Gemini 3.1 Pro by 6.968.561.6+6.9
- AgentWorldBench - OSGemini 3.1 Pro by 6.666.960.3+6.6
- ERQAGemini 3.1 Pro by 5.170.865.7+5.1
- NL2RepoQwen3.6 Plus by 4.533.437.9+4.5
- AgentWorldBench - MCPGemini 3.1 Pro by 3.859.155.3+3.8
- AgentWorldBench - OverallGemini 3.1 Pro by 3.854.650.8+3.8
- AgentWorldBench - AndroidGemini 3.1 Pro by 3.761.457.6+3.7
- τ³-BenchQwen3.6 Plus by 3.667.170.7+3.6
- MMMLUGemini 3.1 Pro by 3.192.689.5+3.1
- AA IntelligenceGemini 3.1 Pro by 2.729.727+2.7
- SimpleVQAGemini 3.1 Pro by 2.669.967.3+2.6
- MMLU-ProGemini 3.1 Pro by 2.59188.5+2.5
- Finance Agent v2Gemini 3.1 Pro by 2.14340.9+2.1
- τ²-Bench Telecom (AA run)Qwen3.6 Plus by 2.195.697.7+2.1
- AgentWorldBench - WebGemini 3.1 Pro by 252.850.8+2
- AgentWorldBench - TerminalGemini 3.1 Pro by 1.952.550.6+1.9
- MATH-VisionGemini 3.1 Pro by 1.889.888+1.8
- HLE (with tools)tie51.450.6tie
- HMMT Nov. 2025tie94.894.6tie
- AgentWorldBench - SWEtie59.159.1tie
- RealWorldQAtie85.485.4tie
Questions people ask
Which is better, Gemini 3.1 Pro or Qwen3.6 Plus?
Gemini 3.1 Pro wins seven of the eight areas we test: coding, reasoning, facts, images and charts, math, long documents and following instructions. Qwen3.6 Plus wins agents, and costs 75% less.
Which is better for coding?
Gemini 3.1 Pro. It wins 7 of the 10 coding tests both models report; Qwen3.6 Plus wins 3.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Qwen3.6 Plus costs $0.50 and $3.00. That makes Qwen3.6 Plus about 75% cheaper for the same work.
How do you compare the two?
We use the 65 benchmark tests both models have published scores on. The verdict counts the 32 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 33 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.