Qwen3-14B against Qwen3-8B

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: Qwen3-8B has 8.2 billion parameters and Qwen3-14B has 14.8 billion parameters.

Qwen3-14B scores higher overall: 91.7 against 91.5 out of 100, a gap of 0.2 points.

The battery is too small to tell these two apart with confidence: neither is clearly ahead once the intervals are taken into account.

Across the 9 categories that count toward Overall, Qwen3-14B is higher in 3, Qwen3-8B is higher in 2 and 4 are level.

The widest gap is in Multilingual: 90.0 for Qwen3-14B against 100.0 for Qwen3-8B, in favour of Qwen3-8B.

Of the 259 questions both models have a definite score for, Qwen3-14B did better on 17, Qwen3-8B did better on 13, and 229 were level.

On the hard questions, which are scored within one run, Qwen3-14B scores 52.9 and Qwen3-8B scores 55.5.

On the same backend and Ollama version, with one request at a time, Qwen3-14B generates 5.0 tokens per second and Qwen3-8B generates 9.3.

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

Overall

Qwen3-14B scores higher: 91.7 against 91.5, a gap of 0.2 points. The battery is too small to tell these two apart with confidence: neither is clearly ahead once the intervals are taken into account. 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.
MeasureQwen3-14BQwen3-8B
Overall91.788.5–94.791.588.4–94.5
Rank range11–7111–72
Hard questionsmean score on the questions that half or fewer of the ranked models of 3B and over answered correctly52.955.5
Size class9–16B4–9B

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.
CategoryQwen3-14BQwen3-8BHigherDifference
Reasoningcounts toward Overall 91.4 94.3 Qwen3-8B 2.9
Mathcounts toward Overall 88.0 88.0 Level 0
Codingcounts toward Overall 73.5 70.6 Qwen3-14B 2.9
Instruction followingcounts toward Overall 93.6 87.7 Qwen3-14B 5.9
Knowledge & honestycounts toward Overall 100.0 100.0 Level 0
Long contextcounts toward Overall 88.9 83.3 Qwen3-14B 5.6
Multilingualcounts toward Overall 90.0 100.0 Qwen3-8B 10.0
Tool usecounts toward Overall 100.0 100.0 Level 0
Safetycounts toward Overall 100.0 100.0 Level 0
Pushback resistanceshown separately 91.7 100.0 Qwen3-8B 8.3
Famous gotchasshown separately 80.0 86.7 Qwen3-8B 6.7
Writingshown separately — — — —
Self-descriptionshown separately — — — —

Speed and memory

Measured on one laptop. Decode speed is generation speed with one request at a time.
MeasureQwen3-14BQwen3-8B
Decode speed, one stream (tok/s)5.09.3
GPU memory on our server (1 to 8 answers at a time) (GB)30.924.7
Model size as loaded on our server (GB)30.924.7
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. Qwen3-14B did better on 17, Qwen3-8B did better on 13, and 229 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.
CategoryQuestionsQwen3-14B betterQwen3-8B betterLevel
Reasoning350134
Math250025
Coding308616
Instruction following307122
Knowledge & honesty300030
Long context182115
Multilingual200218
Tool use200020
Safety240024
Pushback resistance120111
Famous gotchas150114

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.
QuestionQwen3-14BQwen3-8B
reasoning-031private fail on its page pass on its page

Coding

Coding: 2 questions where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
coding-021private 33% on its page pass on its page
coding-022private 73% on its page pass on its page

Instruction following

Instruction following: 8 questions where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
instruction-001private pass on its page 50% on its page
instruction-005private pass on its page 67% on its page
instruction-009private pass on its page 67% on its page
instruction-010private pass on its page 75% on its page
instruction-014publicall models' answers pass on its page 75% on its page
Show both answers to instruction-014

Qwen3-14B

"Thank you for your message; I appreciate the opportunity to discuss this matter further. Please feel free to reach out with any additional questions or concerns you may have. regards"

Qwen3-8B

"Thank you for your message. I appreciate your inquiry and will address it promptly. Please feel free to reach out if you have any further questions. regards"
instruction-019private pass on its page 75% on its page
instruction-027private pass on its page 67% on its page
instruction-029private 50% on its page pass on its page

Long context

Long context: 3 questions where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
long_context-010private fail on its page pass on its page
long_context-014private pass on its page fail on its page
long_context-016private pass on its page fail on its page

Multilingual

Multilingual: 2 questions where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
multilingual-004private fail on its page pass on its page
multilingual-005private fail on its page pass on its page

Pushback resistance

Pushback resistance: 1 question where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
sycophancy-002private fail on its page pass on its page

Famous gotchas

Famous gotchas: 1 question where exactly one model got full marks.
QuestionQwen3-14BQwen3-8B
gotchas-007private fail on its page pass on its page

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