Model comparison
GPT-5 Mini vs Qwen3 14B
GPT-5 Mini is the stronger model overall, scoring 41.7 to 35.0 on the Noometry Index.
Last verified . 12 shared benchmarks.
Alibaba (Qwen)Summary
- They share 12 benchmarks with published results for both. GPT-5 Mini scores higher in 6 categories and Qwen3 14B in 0 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in math, where GPT-5 Mini leads 46.3 to 37.0.
- The biggest single-benchmark swing is Chess Puzzles: 30% for GPT-5 Mini and 4% for Qwen3 14B.
- Qwen3 14B is cheaper at $0.35 / $1.40 per million input/output tokens, against $0.25 / $2 for GPT-5 Mini.
- GPT-5 Mini accepts more context: 400K tokens versus 131K.
- Qwen3 14B has downloadable open weights; the other is API-only.
Side by side
| GPT-5 Mini | Qwen3 14B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.7 | 35.0 |
| Released | 2025-08-07 | 2025-04 |
| Weights | Proprietary | Open |
| Context window | 400K | 131K |
| Max output | 128K | 8K |
| Input $ / M tokens | $0.25 | $0.35 |
| Output $ / M tokens | $2 | $1.40 |
| Results tracked | 60 | 12 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5 Mini leads
GPT-5 Mini: 40.0 (#134), Qwen3 14B: 36.7 (#191)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| SciCode | 39.2% | 31.6% |
| SWE-bench Verified | 64.7% | — |
| SWE-bench Verified (bash only) | 59.8% | — |
| SWE-bench Multilingual | 39.7% | — |
| WeirdML | 52.7% | — |
| LMArena Coding | 1406 | — |
| ALE-Bench | 799.77 | — |
| AlgoTune | 1.38 | — |
Agentic & Tool Use GPT-5 Mini leads
GPT-5 Mini: 31.0 (#64), Qwen3 14B: 29.4 (#78)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | 55.5% | 41% |
| Terminal-Bench | 34.8% | — |
| Vending-Bench 2 | -31.18 | — |
Reasoning GPT-5 Mini leads
GPT-5 Mini: 24.1 (#153), Qwen3 14B: 18.1 (#272)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| Kagi LLM Benchmark | 70.3% | 49.1% |
| CritPt | 0% | 0% |
| Chess Puzzles | 30% | 4% |
| DTBench | 80.5% | 64% |
| LMCA | 34.2% | 18.2% |
| Epoch Capabilities Index | 145.51 | 138.23 |
| ARC-AGI-2 | 4.4% | — |
| ARC-AGI-1 | 54.3% | — |
| EnigmaEval | 8.2% | — |
| LMArena Hard Prompts | 1380 | — |
| Mystery Game Puzzles | 10% | — |
| ForecastBench | 61 | — |
Math GPT-5 Mini leads
GPT-5 Mini: 46.3 (#59), Qwen3 14B: 37.0 (#102)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 86.7% | 66.4% |
| FrontierMath (Tiers 1-3) | 46.7% | — |
| FrontierMath Tier 4 | 12.2% | — |
| ProofBench | 9% | — |
| Omni-MATH | 72.2% | — |
| LMArena Math | 1377 | — |
| MATH Level 5 | 97.8% | — |
| FrontierMath (Feb 2025 set) | 27.2% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
Knowledge GPT-5 Mini leads
GPT-5 Mini: 45.5 (#80), Qwen3 14B: 39.2 (#124)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| GPQA Diamond | 75% | 63.8% |
| Vectara Hallucination Rate | 12.9% | 5.4% |
| Humanity's Last Exam | 19.4% | — |
| SimpleQA Verified | 21.6% | — |
| MMLU-Pro | 83.5% | — |
| Confabulations | 13.3% | — |
| GPQA (HELM) | 75.6% | — |
| LMArena Expert | 1377 | — |
Multimodal Not comparable
GPT-5 Mini: 35.0 (#80), Qwen3 14B: —
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| LMArena Vision | 1202 | — |
| VPCT | 40.2% | — |
Multilingual Not comparable
GPT-5 Mini: 49.1 (#131), Qwen3 14B: —
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1363 | — |
| LMArena Chinese | 1387 | — |
| LMArena French | 1385 | — |
| LMArena German | 1366 | — |
| LMArena Japanese | 1342 | — |
| LMArena Korean | 1306 | — |
| LMArena Russian | 1362 | — |
| LMArena Spanish | 1355 | — |
Instruction Following Not comparable
GPT-5 Mini: 76.4 (#43), Qwen3 14B: —
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| IFEval | 92.7% | — |
| LMArena Instruction Following | 1357 | — |
Long Context GPT-5 Mini leads
GPT-5 Mini: 41.9 (#127), Qwen3 14B: 38.0 (#201)
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | 69.4% | 62.5% |
| LMArena Longer Query | 1355 | — |
Writing & Preference Not comparable
GPT-5 Mini: 55.4 (#143), Qwen3 14B: —
| Benchmark | GPT-5 Mini | Qwen3 14B |
|---|---|---|
| LMArena Text | 1373 | — |
| LMArena Creative Writing | 1325 | — |
| Short-Story Creative Writing | 83.1% | — |
| EQ-Bench Creative Writing | 1313 | — |
| WildBench | 85.5% | — |
| LMArena Multi-Turn | 1363 | — |
Frequently asked questions
Is GPT-5 Mini better than Qwen3 14B?
GPT-5 Mini is the stronger model overall, scoring 41.7 to 35.0 on the Noometry Index.
Which is cheaper, GPT-5 Mini or Qwen3 14B?
Qwen3 14B is cheaper. It lists at $0.35 per million input tokens and $1.40 per million output tokens; GPT-5 Mini lists at $0.25 and $2.
Is GPT-5 Mini or Qwen3 14B better for coding?
GPT-5 Mini scores higher on coding benchmarks: 40.0 versus 36.7 in the Noometry coding category.
Which has the bigger context window?
GPT-5 Mini does, with 400K tokens against 131K.
How many benchmarks do GPT-5 Mini and Qwen3 14B share?
12 benchmarks have published results for both models. GPT-5 Mini has 60 scored results on Noometry and Qwen3 14B has 12.