Qwen3.5 27B vs Qwen3.5 2B
Wins 6 of 7 areas
Coding · Agents · Reasoning · Images and charts · Long documents · Following instructions
Wins 0 of 7 areas
—
Qwen3.5 27B is the stronger all-rounder.
Scores updated · 34 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.
- Images and chartsUnderstanding pictures, charts and video40Qwen3.5 27B4 of 4 tests
- ReasoningHard problems that need careful thinking20Qwen3.5 27B2 of 3 tests · 1 tie
- CodingWriting and fixing software20Qwen3.5 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
- AgentsCarrying out multi-step tasks on its own10Qwen3.5 27B1 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Agents 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.
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
- Reasons across sets of long documentsAA-LCR+56.7points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+44.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+40.2points ahead
Where Qwen3.5 2B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+7.3points ahead
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingQwen3.5 27B
- SciCodeQwen3.5 27B by 36.739.52.8+36.7
- Terminal-Bench HardQwen3.5 27B by 28.832.63.8+28.8
ReasoningQwen3.5 27B
- GPQA DiamondQwen3.5 27B by 40.285.845.6+40.2
- Humanity's Last ExamQwen3.5 27B by 21.323.92.6+21.3
- CritPttie0.90tie
FactsEven
- AA-Omniscience · AccuracyQwen3.5 27B by 12.620.78.1+12.6
- AA-Omniscience · Non-hallucinationQwen3.5 2B by 7.318.525.8+7.3
Images and chartsQwen3.5 27B
Long documentsQwen3.5 27B
- AA-LCRQwen3.5 27B by 56.777.721+56.7
- LongBench v2Qwen3.5 27B by 21.960.638.7+21.9
Following instructionsQwen3.5 27B
- IFBenchQwen3.5 27B by 44.175.631.5+44.1
- Multi-ChallengeQwen3.5 27B by 27.160.833.7+27.1
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.
- Artificial Analysis Coding IndexQwen3.5 27B by 3234.92.9+32
- WMT24++Qwen3.5 27B by 31.877.645.8+31.8
- t2-benchQwen3.5 27B by 30.27948.8+30.2
- MMLU-ProXQwen3.5 27B by 29.982.252.3+29.9
- SuperGPQAQwen3.5 27B by 28.165.637.5+28.1
- IncludeQwen3.5 27B by 26.281.655.4+26.2
- BFCLv4Qwen3.5 27B by 24.968.543.6+24.9
- τ²-Bench Telecom (AA run)Qwen3.5 27B by 24.993.969+24.9
- MMMLUQwen3.5 27B by 22.885.963.1+22.8
- SimpleVQAQwen3.5 27B by 20.85635.2+20.8
- MMLU-ProQwen3.5 27B by 19.686.166.5+19.6
- RealWorldQAQwen3.5 27B by 18.683.765.1+18.6
- C-EvalQwen3.5 27B by 17.390.573.2+17.3
- IFEvalQwen3.5 27B by 16.49578.6+16.4
- AA-OmniscienceQwen3.5 27B by 16.2-44-60.2+16.2
- AA IntelligenceQwen3.5 27B by 1622.96.9+16
- MMLU-ReduxQwen3.5 27B by 13.693.279.6+13.6
- HallusionBenchQwen3.5 27B by 4.57065.5+4.5
Questions people ask
Which is better, Qwen3.5 27B or Qwen3.5 2B?
Qwen3.5 27B wins six of the seven areas where both have results: coding, agents, reasoning, images and charts, long documents and following instructions. Qwen3.5 2B wins none. They are level on facts.
Which is better for coding?
Qwen3.5 27B. It wins 2 of the 2 coding tests both models report; Qwen3.5 2B wins none.
How do you compare the two?
We use the 34 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 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.