Gemini 3.1 Pro vs Qwen3.8 27B
Wins 3 of 8 areas
Reasoning · Images and charts · Math
Wins 2 of 8 areas
Coding · Agents
Gemini 3.1 Pro wins more areas, narrowly.Qwen3.8 27B is cheaper and better at agents.
Both rank among the ten best models we track in following instructions.
Scores updated · 42 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.
- ReasoningHard problems that need careful thinking60Gemini 3.1 Pro6 of 6 tests
- Images and chartsUnderstanding pictures, charts and video21Gemini 3.1 Pro2 of 4 tests · 1 tie
- MathCompetition and research-level math10Gemini 3.1 Pro1 of 1 test
- AgentsCarrying out multi-step tasks on its own04Qwen3.8 27B4 of 4 tests
- CodingWriting and fixing software34Qwen3.8 27B4 of 8 tests · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
- Following instructionsDoing exactly what it is asked11Even1 each
Math and long documents 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.
Qwen3.8 27B 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
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+39.3points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+34.7points ahead
- Common-sense trick questionsSimpleBench+19.4points ahead
Where Qwen3.8 27B pulls ahead
- Real work tasks from 44 professionsGDPVal+31.5points ahead
- Customer-service tasks in a simulated bankτ-Bench V3 · Banking+26.6points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+20.6points ahead
Every test, side by side
All 42 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.8 27B
- LiveBench · Agentic CodingQwen3.8 27B by 17.344.161.4+17.3
- LMArena · WebDevQwen3.8 27B by 145 rating points14471592+145 rating
- SciCodeGemini 3.1 Pro by 12.158.746.6+12.1
- SWE-bench ProQwen3.8 27B by 7.554.261.7+7.5
- Terminal-Bench 2.1Qwen3.8 27B by 673.879.8+6
- SWE-bench MultilingualGemini 3.1 Pro by 3.176.973.8+3.1
- LiveCodeBench v6Gemini 3.1 Pro by 1.491.790.3+1.4
- LiveBench · Codingtie76.575.7tie
AgentsQwen3.8 27B
- GDPValQwen3.8 27B by 31.514.746.2+31.5
- τ-Bench V3 · BankingQwen3.8 27B by 26.621.448+26.6
- OSWorld-VerifiedQwen3.8 27B by 8.176.284.3+8.1
- Terminal-Bench 4.0Qwen3.8 27B by 1.645.6+1.6
ReasoningGemini 3.1 Pro
- ARC-AGI-2Gemini 3.1 Pro by 34.777.142.4+34.7
- SimpleBenchGemini 3.1 Pro by 19.479.660.2+19.4
- Humanity's Last ExamGemini 3.1 Pro by 13.14733.9+13.1
- CritPtGemini 3.1 Pro by 12.317.75.4+12.3
- LiveBench · ReasoningGemini 3.1 Pro by 48480+4
- GPQA DiamondGemini 3.1 Pro by 3.694.190.5+3.6
FactsEven
- AA-Omniscience · AccuracyGemini 3.1 Pro by 39.354.915.6+39.3
- AA-Omniscience · Non-hallucinationQwen3.8 27B by 20.649.169.7+20.6
Images and chartsGemini 3.1 Pro
- CharXiv (RQ)Qwen3.8 27B by 6.983.390.2+6.9
- MMMU-ProGemini 3.1 Pro by 6.182.476.3+6.1
- LMArena · VisionGemini 3.1 Pro by 26 rating points12961270+26 rating
- OCRBenchv2tie62.863.5tie
Following instructionsEven
- LiveBench · Instruction FollowingGemini 3.1 Pro by 6.479.172.7+6.4
- IFBenchQwen3.8 27B by 2.477.179.5+2.4
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 3.1 Pro by 41.931.9-10+41.9
- AA Agentic IndexQwen3.8 27B by 36.210.346.5+36.2
- BabyVisionQwen3.8 27B by 31.254.485.6+31.2
- DeepSWE 1.1Qwen3.8 27B by 30.21242.2+30.2
- livebench_languageGemini 3.1 Pro by 1185.474.3+11
- ARC-AGI-1Gemini 3.1 Pro by 10.59887.5+10.5
- NL2RepoQwen3.8 27B by 8.933.442.3+8.9
- LVBenchQwen3.8 27B by 6.266.272.4+6.2
- ERQAGemini 3.1 Pro by 5.370.865.5+5.3
- MATH-VisionQwen3.8 27B by 4.889.894.6+4.8
- AA IntelligenceQwen3.8 27B by 429.733.7+4
- livebench_data_analysisGemini 3.1 Pro by 1.978.576.6+1.9
- Artificial Analysis Coding Indextie68.868.1tie
- RealWorldQAtie85.485.9tie
Questions people ask
Which is better, Gemini 3.1 Pro or Qwen3.8 27B?
Gemini 3.1 Pro wins three of the eight areas we test: reasoning, images and charts and math. Qwen3.8 27B wins coding and agents, and costs 75% less. They are level on facts, long documents and following instructions.
Which is better for coding?
Qwen3.8 27B. It wins 4 of the 8 coding tests both models report; Gemini 3.1 Pro wins 3, and 1 is a tie.
Which is cheaper?
Gemini 3.1 Pro costs $2.00 per million input tokens and $12.00 per million output tokens; Qwen3.8 27B costs $0.50 and $3.00. That makes Qwen3.8 27B about 75% cheaper for the same work.
How do you compare the two?
We use the 42 benchmark tests both models have published scores on. The verdict counts the 28 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.