Gemini 3.1 Pro vs Qwen3.7 Max
Wins 3 of 7 areas
Reasoning · Facts · Long documents
Wins 2 of 7 areas
Agents · Math
Gemini 3.1 Pro wins more areas, narrowly.Qwen3.7 Max is cheaper and better at math.
Both rank among the ten best models we track in following instructions.
Scores updated · 50 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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 3.1 Pro4 of 5 tests · 1 tie
- 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
- MathCompetition and research-level math24Qwen3.7 Max4 of 6 tests
- AgentsCarrying out multi-step tasks on its own23Qwen3.7 Max3 of 5 tests
- CodingWriting and fixing software33Even3 each · 4 ties
- Following instructionsDoing exactly what it is asked11Even1 each
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.
Qwen3.7 Max costs 29% 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
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+23.8points ahead
- Short factual questions, answered correctlySimpleQA Verified+17.7points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+9.6points ahead
Where Qwen3.7 Max pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+25.3points ahead
- Real work tasks from 44 professionsGDPVal+16.9points ahead
- Finds the root cause of IT system incidentsAA IT-Bench SRE+12.2points ahead
Every test, side by side
All 50 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- SciCodeGemini 3.1 Pro by 9.258.749.5+9.2
- LMArena · WebDevQwen3.7 Max by 69 rating points14461515+69 rating
- SWE-bench ProQwen3.7 Max by 6.454.260.6+6.4
- Terminal-Bench HardGemini 3.1 Pro by 353.850.8+3
- LiveBench · CodingGemini 3.1 Pro by 2.376.574.2+2.3
- SWE-bench MultilingualQwen3.7 Max by 1.476.978.3+1.4
- Terminal-Bench 2.1tie73.874.5tie
- LiveBench · Agentic Codingtie44.143.6tie
- SWE-bench Verifiedtie80.680.4tie
- LiveCodeBench v6tie91.791.6tie
AgentsQwen3.7 Max
- GDPValQwen3.7 Max by 16.914.731.6+16.9
- AA IT-Bench SREQwen3.7 Max by 12.230.342.5+12.2
- τ-Bench V3 · BankingGemini 3.1 Pro by 9.621.411.8+9.6
- MCP AtlasQwen3.7 Max by 7.269.276.4+7.2
- Terminal-Bench 4.0Gemini 3.1 Pro by 2.541.5+2.5
ReasoningGemini 3.1 Pro
- SimpleBenchGemini 3.1 Pro by 9.279.670.4+9.2
- Humanity's Last ExamGemini 3.1 Pro by 6.54740.5+6.5
- CritPtGemini 3.1 Pro by 4.317.713.4+4.3
- GPQA DiamondGemini 3.1 Pro by 1.894.192.3+1.8
- LiveBench · Reasoningtie8483.3tie
FactsGemini 3.1 Pro
- AA-Omniscience · Non-hallucinationQwen3.7 Max by 25.349.174.4+25.3
- AA-Omniscience · AccuracyGemini 3.1 Pro by 23.854.931.1+23.8
- SimpleQA VerifiedGemini 3.1 Pro by 17.773.555.8+17.7
MathQwen3.7 Max
- IMOAnswerBenchQwen3.7 Max by 98190+9
- FrontierMath Tier 4Qwen3.7 Max by 7.426.834.1+7.4
- LiveBench · MathematicsGemini 3.1 Pro by 5.79185.3+5.7
- FrontierMath Tiers 1-3 (v2)Qwen3.7 Max by 4.959.664.6+4.9
- HMMT Feb. 2026Qwen3.7 Max by 2.494.797.1+2.4
- AIME 2026Gemini 3.1 Pro by 1.398.397+1.3
Following instructionsEven
- LiveBench · Instruction FollowingGemini 3.1 Pro by 5.179.174+5.1
- IFBenchQwen3.7 Max by 3.477.180.5+3.4
Other results18 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 18.431.913.5+18.4
- NL2RepoQwen3.7 Max by 13.833.447.2+13.8
- AA Agentic IndexQwen3.7 Max by 13.610.323.9+13.6
- DeepSWE 1.1Qwen3.7 Max by 9.61221.6+9.6
- DeepSWEQwen3.7 Max by 81018+8
- livebench_data_analysisGemini 3.1 Pro by 6.778.571.8+6.7
- livebench_languageGemini 3.1 Pro by 5.785.479.7+5.7
- LiveBenchGemini 3.1 Pro by 5.679.974.3+5.6
- Finance Agent v2Qwen3.7 Max by 5.44348.4+5.4
- MCPMarkQwen3.7 Max by 4.955.960.8+4.9
- Artificial Analysis Coding IndexGemini 3.1 Pro by 2.868.866+2.8
- MMMLUGemini 3.1 Pro by 2.392.690.3+2.3
- HLE (with tools)Qwen3.7 Max by 2.151.453.5+2.1
- MMLU-ProGemini 3.1 Pro by 1.49189.6+1.4
- Terminal-Bench 2.0Qwen3.7 Max by 1.268.569.7+1.2
- τ²-Bench Telecom (AA run)tie95.694.7tie
- AA Intelligencetie29.729.5tie
- HMMT Nov. 2025tie94.895tie
Questions people ask
Which is better, Gemini 3.1 Pro or Qwen3.7 Max?
Gemini 3.1 Pro wins three of the seven areas where both have results: reasoning, facts and long documents. Qwen3.7 Max wins agents and math, and costs 29% less. They are level on coding and following instructions.
Which is better for coding?
Neither. They win 3 coding tests each of the 10 both models report, and 4 are ties.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Qwen3.7 Max costs $2.50 and $7.50. That makes Qwen3.7 Max about 29% cheaper for the same work.
How do you compare the two?
We use the 50 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 18 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.