Qwen3.5 27B vs Qwen3.5 9B
Wins 8 of 8 areas
Coding · Agents · Reasoning · Facts · Images and charts · Math · Long documents · Following instructions
Wins 0 of 8 areas
—
Qwen3.5 27B is the stronger all-rounder.Qwen3.5 9B is cheaper.
Scores updated · 37 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 software30Qwen3.5 27B3 of 3 tests
- ReasoningHard problems that need careful thinking20Qwen3.5 27B2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up20Qwen3.5 27B2 of 2 tests
- Long documentsFinding answers in very long texts20Qwen3.5 27B2 of 2 tests
- Following instructionsDoing exactly what it is asked20Qwen3.5 27B2 of 2 tests
- MathCompetition and research-level math10Qwen3.5 27B1 of 2 tests · 1 tie
- AgentsCarrying out multi-step tasks on its own10Qwen3.5 27B1 of 1 test
- Images and chartsUnderstanding pictures, charts and video10Qwen3.5 27B1 of 1 test
Agents and images and charts 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.
Qwen3.5 9B costs 88% 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 Qwen3.5 27B pulls ahead
- Recent programming contest problemsLiveCodeBench v6+15.1points ahead
- Code for real scientific research problemsSciCode+10points ahead
- Harvard-MIT high-school math contest problemsHMMT Feb. 2026+9.9points ahead
Where Qwen3.5 9B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 37 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.5 27B
- LiveCodeBench v6Qwen3.5 27B by 15.180.765.6+15.1
- SciCodeQwen3.5 27B by 1039.529.5+10
- Terminal-Bench HardQwen3.5 27B by 8.432.624.2+8.4
ReasoningQwen3.5 27B
- Humanity's Last ExamQwen3.5 27B by 923.914.9+9
- GPQA DiamondQwen3.5 27B by 5.285.880.6+5.2
- CritPttie0.90.3tie
FactsQwen3.5 27B
- AA-Omniscience · AccuracyQwen3.5 27B by 4.320.716.4+4.3
- AA-Omniscience · Non-hallucinationQwen3.5 27B by 2.118.516.4+2.1
Long documentsQwen3.5 27B
- AA-LCRQwen3.5 27B by 7.777.770+7.7
- LongBench v2Qwen3.5 27B by 5.460.655.2+5.4
Following instructionsQwen3.5 27B
- IFBenchQwen3.5 27B by 8.975.666.7+8.9
- Multi-ChallengeQwen3.5 27B by 6.360.854.5+6.3
Other results21 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 Agentic IndexQwen3.5 27B by 47.654.67+47.6
- MLVUQwen3.5 27B by 33.585.952.4+33.5
- VITA-BenchQwen3.5 27B by 12.141.929.8+12.1
- AA IntelligenceQwen3.5 27B by 11.722.911.2+11.7
- AA-OmniscienceQwen3.5 27B by 9.5-44-53.5+9.5
- BFCLv4Qwen3.5 27B by 8.468.560.1+8.4
- SuperGPQAQwen3.5 27B by 7.465.658.2+7.4
- τ²-Bench Telecom (AA run)Qwen3.5 27B by 7.193.986.8+7.1
- HMMT 2025Qwen3.5 27B by 6.989.882.9+6.9
- Artificial Analysis Coding IndexQwen3.5 27B by 6.234.928.7+6.2
- IncludeQwen3.5 27B by 681.675.6+6
- MMLU-ProXQwen3.5 27B by 5.982.276.3+5.9
- MMMLUQwen3.5 27B by 4.785.981.2+4.7
- MMVUQwen3.5 27B by 4.773.368.6+4.7
- DeepPlanningQwen3.5 27B by 4.622.618+4.6
- MMLU-ProQwen3.5 27B by 3.686.182.5+3.6
- IFEvalQwen3.5 27B by 3.59591.5+3.5
- C-EvalQwen3.5 27B by 2.390.588.2+2.3
- Claw-Eval AvgQwen3.5 9B by 2.264.366.5+2.2
- MMLU-ReduxQwen3.5 27B by 2.193.291.1+2.1
- t2-benchtie7979.1tie
Questions people ask
Which is better, Qwen3.5 27B or Qwen3.5 9B?
Qwen3.5 27B wins all eight areas we test: coding, agents, reasoning, facts, images and charts, math, long documents and following instructions. Qwen3.5 9B wins none, but costs 88% less.
Which is better for coding?
Qwen3.5 27B. It wins 3 of the 3 coding tests both models report; Qwen3.5 9B wins none.
Which is cheaper?
Qwen3.5 27B costs $0.30 per million input tokens and $2.40 per million output tokens; Qwen3.5 9B costs $0.14 and $0.20. That makes Qwen3.5 9B about 88% cheaper for the same work.
How do you compare the two?
We use the 37 benchmark tests both models have published scores on. The verdict counts the 16 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 21 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.