Qwen3.5-27B against Qwen3.5-9B

Pick two models to see them side by side: Overall score, every category, the questions where one did better, and speed and memory. Each model is shown at its newest tested version. Back to the leaderboard.

The short version

Same model line in two sizes: Qwen3.5-9B has 9.7 billion parameters and Qwen3.5-27B has 27.8 billion parameters.

Qwen3.5-27B scores higher overall: 96.2 against 92.0 out of 100, a gap of 4.2 points.

With the 95% intervals taken into account, Qwen3.5-27B is clearly ahead of Qwen3.5-9B.

Across the 9 categories that count toward Overall, Qwen3.5-27B is higher in 5, Qwen3.5-9B is higher in 0 and 4 are level.

The widest gap is in Coding: 93.1 for Qwen3.5-27B against 77.4 for Qwen3.5-9B, in favour of Qwen3.5-27B.

Of the 259 questions both models have a definite score for, Qwen3.5-27B did better on 20, Qwen3.5-9B did better on 3, and 236 were level.

On the hard questions, which are scored within one run, Qwen3.5-27B scores 78.8 and Qwen3.5-9B scores 59.9.

On the same backend and Ollama version, with one request at a time, Qwen3.5-27B generates 2.7 tokens per second and Qwen3.5-9B generates 8.4.

Written from the numbers below. Scores come from one fixed battery run on one laptop; read the methodology before drawing conclusions.

Overall

Qwen3.5-27B scores higher: 96.2 against 92.0, a gap of 4.2 points. With the 95% intervals taken into account, Qwen3.5-27B is clearly ahead of Qwen3.5-9B. How sure is a score?

Overall (0–100) with its 95% interval, the rank range among all listed ranked models, and the hard-question score.
MeasureQwen3.5-27BQwen3.5-9B
Overall96.293.8–98.392.089.0–95.1
Rank range3–3511–69
Hard questionsmean score on the questions that half or fewer of the ranked models of 3B and over answered correctly78.859.9
Size class16–35B9–16B

By category

Programmatic mean (0–100) per category. Higher is the model with the larger mean; a gap that rounds to 0.0 is called level. A dash means not scored, which is different from 0.
CategoryQwen3.5-27BQwen3.5-9BHigherDifference
Reasoningcounts toward Overall 88.6 85.7 Qwen3.5-27B 2.9
Mathcounts toward Overall 84.0 84.0 Level 0
Codingcounts toward Overall 93.1 77.4 Qwen3.5-27B 15.7
Instruction followingcounts toward Overall 100.0 94.1 Qwen3.5-27B 5.9
Knowledge & honestycounts toward Overall 100.0 100.0 Level 0
Long contextcounts toward Overall 100.0 88.9 Qwen3.5-27B 11.1
Multilingualcounts toward Overall 100.0 98.3 Qwen3.5-27B 1.7
Tool usecounts toward Overall 100.0 100.0 Level 0
Safetycounts toward Overall 100.0 100.0 Level 0
Pushback resistanceshown separately 100.0 100.0 Level 0
Famous gotchasshown separately 93.3 93.3 Level 0
Writingshown separately — — — —
Self-descriptionshown separately — — — —

Speed and memory

Measured on one laptop. Decode speed is generation speed with one request at a time.
MeasureQwen3.5-27BQwen3.5-9B
Decode speed, one stream (tok/s)2.78.4
GPU memory on our server (1 to 8 answers at a time) (GB)17.66.1
Model size as loaded on our server (GB)17.66.1
BackendROCm 7.1 (Ollama 0.34.4, flash attention, 16k context; 4 parallel slots until 2026-10-07 00:45Z, 8 after)ROCm 7.1 (Ollama 0.34.4, flash attention, 16k context; 4 parallel slots until 2026-10-07 00:45Z, 8 after)

A dash means the figure was not measured, which is different from 0. Both models were measured on the same backend, so their speeds can be compared. Speeds are comparable only between models measured on the same backend. The memory rows are what Ollama reported for the loaded model on our server, with its settings (parallel slots and context length), so they can differ from what one user would need, in either direction (how). They are not the memory estimate the hardware picker uses. See also the Speed page.

Where they differ

Both models have a definite score for 259 questions. Qwen3.5-27B did better on 20, Qwen3.5-9B did better on 3, and 236 were level. A score gap of more than 0.001 counts as a win. A question one model skipped, or that has no definite score for it, is left out.

Questions both models have a score for, by category.
CategoryQuestionsQwen3.5-27B betterQwen3.5-9B betterLevel
Reasoning351034
Math251123
Coding309219
Instruction following306024
Knowledge & honesty300030
Long context182016
Multilingual201019
Tool use200020
Safety240024
Pushback resistance120012
Famous gotchas150015

Questions where only one got full marks

Full marks means a score of 1. The answers of public questions can be opened below. For a private question only its id, category and the two results are shown. Each result links to that question on the model's page.

Reasoning

Reasoning: 1 question where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
reasoning-035private pass on its page fail on its page

Math

Math: 2 questions where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
math-004private pass on its page fail on its page
math-025private fail on its page pass on its page

Coding

Coding: 8 questions where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
coding-006private pass on its page fail on its page
coding-011private pass on its page 10% on its page
coding-017private pass on its page 40% on its page
coding-019private pass on its page 63% on its page
coding-022private pass on its page 27% on its page
coding-023private 89% on its page pass on its page
coding-024private pass on its page fail on its page
coding-030private pass on its page 91% on its page

Instruction following

Instruction following: 6 questions where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
instruction-009private pass on its page 67% on its page
instruction-016private pass on its page 67% on its page
instruction-018private pass on its page 80% on its page
instruction-021private pass on its page 75% on its page
instruction-027private pass on its page 67% on its page
instruction-028private pass on its page 67% on its page

Long context

Long context: 2 questions where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
long_context-005private pass on its page fail on its page
long_context-008private pass on its page fail on its page

Multilingual

Multilingual: 1 question where exactly one model got full marks.
QuestionQwen3.5-27BQwen3.5-9B
multilingual-020private pass on its page 67% on its page

The same numbers as JSON: compare.json. Answers appear there only for public questions.