Gemini 3.1 Pro vs Qwen3.5 35B A3B
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
Gemini 3.1 Pro is the stronger all-rounder.Qwen3.5 35B A3B is cheaper.
Both rank among the ten best models we track in images and charts.
Scores updated · 64 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 software80Gemini 3.1 Pro8 of 8 tests
- AgentsCarrying out multi-step tasks on its own60Gemini 3.1 Pro6 of 6 tests
- ReasoningHard problems that need careful thinking30Gemini 3.1 Pro3 of 3 tests
- MathCompetition and research-level math30Gemini 3.1 Pro3 of 3 tests
- Images and chartsUnderstanding pictures, charts and video31Gemini 3.1 Pro3 of 4 tests
- FactsGetting facts right instead of making them up20Gemini 3.1 Pro2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked20Gemini 3.1 Pro2 of 2 tests
- Long documentsFinding answers in very long texts10Gemini 3.1 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.
Qwen3.5 35B A3B costs 84% 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 3.1 Pro pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+34.8points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+34.5points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+33points ahead
Where Qwen3.5 35B A3B pulls ahead
- Reads and reasons about text in imagesOCRBenchv2+2.5points ahead
Every test, side by side
All 64 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemini 3.1 Pro
- Terminal-Bench 2.1Gemini 3.1 Pro by 3373.840.8+33
- Terminal-Bench HardGemini 3.1 Pro by 27.353.826.5+27.3
- SciCodeGemini 3.1 Pro by 2158.737.7+21
- LMArena · WebDevGemini 3.1 Pro by 197 rating points14471250+197 rating
- LiveCodeBench v6Gemini 3.1 Pro by 17.191.774.6+17.1
- SWE-bench MultilingualGemini 3.1 Pro by 16.676.960.3+16.6
- SWE-bench VerifiedGemini 3.1 Pro by 11.480.669.2+11.4
- SWE-bench ProGemini 3.1 Pro by 9.654.244.6+9.6
AgentsGemini 3.1 Pro
- BrowseCompGemini 3.1 Pro by 24.985.961+24.9
- OSWorld-VerifiedGemini 3.1 Pro by 21.776.254.5+21.7
- τ-Bench V3 · BankingGemini 3.1 Pro by 16.521.44.9+16.5
- GDPValGemini 3.1 Pro by 9.114.75.6+9.1
- AA IT-Bench SREGemini 3.1 Pro by 8.830.321.5+8.8
- MCP AtlasGemini 3.1 Pro by 6.869.262.4+6.8
ReasoningGemini 3.1 Pro
- Humanity's Last ExamGemini 3.1 Pro by 264721+26
- CritPtGemini 3.1 Pro by 16.817.70.9+16.8
- GPQA DiamondGemini 3.1 Pro by 9.694.184.5+9.6
FactsGemini 3.1 Pro
- AA-Omniscience · AccuracyGemini 3.1 Pro by 34.854.920.1+34.8
- AA-Omniscience · Non-hallucinationGemini 3.1 Pro by 34.549.114.6+34.5
- Vectara HHEM hallucination ratelower is bettertie10.410.5tie
Images and chartsGemini 3.1 Pro
- MMMU-ProGemini 3.1 Pro by 9.782.472.7+9.7
- CharXiv (RQ)Gemini 3.1 Pro by 5.883.377.5+5.8
- MathVistaGemini 3.1 Pro by 490.286.2+4
- OCRBenchv2Qwen3.5 35B A3B by 2.562.865.3+2.5
MathGemini 3.1 Pro
- HMMT Feb. 2026Gemini 3.1 Pro by 12.994.781.8+12.9
- AIME 2026Gemini 3.1 Pro by 598.393.3+5
- IMOAnswerBenchGemini 3.1 Pro by 4.28176.8+4.2
Following instructionsGemini 3.1 Pro
- Multi-ChallengeGemini 3.1 Pro by 11.471.460+11.4
- IFBenchGemini 3.1 Pro by 4.677.172.5+4.6
Other results34 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 8031.9-48.1+80
- Artificial Analysis Coding IndexGemini 3.1 Pro by 31.868.837+31.8
- MCPMarkGemini 3.1 Pro by 28.955.927+28.9
- Terminal-Bench 2.0Gemini 3.1 Pro by 2868.540.5+28
- Tool-DecathlonGemini 3.1 Pro by 20.148.828.7+20.1
- t2-benchGemini 3.1 Pro by 18.199.381.2+18.1
- BabyVisionGemini 3.1 Pro by 1654.438.4+16
- DynaMathQwen3.5 35B A3B by 12.972.185+12.9
- NL2RepoGemini 3.1 Pro by 12.933.420.5+12.9
- SimpleVQAGemini 3.1 Pro by 11.669.958.3+11.6
- AgentWorldBench - SWEGemini 3.1 Pro by 11.559.147.6+11.5
- MMLUGemini 3.1 Pro by 11.592.681.1+11.5
- AgentWorldBench - OSGemini 3.1 Pro by 10.666.956.3+10.6
- AA IntelligenceGemini 3.1 Pro by 10.429.719.3+10.4
- AgentWorldBench - AndroidGemini 3.1 Pro by 8.261.453.2+8.2
- MMMLUGemini 3.1 Pro by 7.492.685.2+7.4
- AgentWorldBench - OverallGemini 3.1 Pro by 6.954.647.7+6.9
- AgentWorldBench - TerminalGemini 3.1 Pro by 6.452.546.1+6.4
- τ²-Bench Telecom (AA run)Gemini 3.1 Pro by 6.495.689.2+6.4
- ERQAGemini 3.1 Pro by 670.864.8+6
- MATH-VisionGemini 3.1 Pro by 5.989.883.9+5.9
- AgentWorldBench - WebGemini 3.1 Pro by 5.752.847.1+5.7
- MMLU-ProGemini 3.1 Pro by 5.79185.3+5.7
- HMMT Nov. 2025Gemini 3.1 Pro by 5.694.889.2+5.6
- LVBenchQwen3.5 35B A3B by 5.266.271.4+5.2
- AgentWorldBench - SearchGemini 3.1 Pro by 4.230.226+4.2
- τ³-BenchQwen3.5 35B A3B by 1.867.168.9+1.8
- AA Agentic IndexQwen3.5 35B A3B by 1.510.311.8+1.5
- RealWorldQAGemini 3.1 Pro by 1.385.484.1+1.3
- vectara_avg_summary_lengthGemini 3.1 Pro by 12.8 rating points107.794.9+12.8 rating
- AgentWorldBench - MCPGemini 3.1 Pro by 1.259.157.9+1.2
- EmbSpatial-BenchGemini 3.1 Pro by 1.184.283.1+1.1
- vectara_answer_ratetie99.499.8tie
- vectara_factual_consistencytie89.689.5tie
Questions people ask
Which is better, Gemini 3.1 Pro or Qwen3.5 35B A3B?
Gemini 3.1 Pro wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Qwen3.5 35B A3B wins none, but costs 84% less.
Which is better for coding?
Gemini 3.1 Pro. It wins 8 of the 8 coding tests both models report; Qwen3.5 35B A3B wins none.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Qwen3.5 35B A3B costs $0.25 and $2.00. That makes Qwen3.5 35B A3B about 84% cheaper for the same work.
How do you compare the two?
We use the 64 benchmark tests both models have published scores on. The verdict counts the 30 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 34 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.