Model comparison
GPT-5.4 nano vs Qwen-14B
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 31.4 on the Noometry Index.
Last verified . 11 shared benchmarks.
Alibaba (Qwen)Summary
- They share 11 benchmarks with published results for both. GPT-5.4 nano scores higher in 7 categories and Qwen-14B in 0 categories; 7 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where GPT-5.4 nano leads 55.7 to 27.6.
- Qwen-14B has downloadable open weights; the other is API-only.
Side by side
| GPT-5.4 nano | Qwen-14B | |
|---|---|---|
| Provider | OpenAI | Alibaba (Qwen) |
| Noometry Index | 41.9 | 31.4 |
| Released | 2026-03-17 | 2023-09-24 |
| Weights | Proprietary | Open |
| Context window | 400K | — |
| Max output | 128K | — |
| Input $ / M tokens | $0.20 | — |
| Output $ / M tokens | $1.25 | — |
| Results tracked | 40 | 18 |
Sponsored placements are available on pages like this one. Advertise on Noometry
Category by category
Coding GPT-5.4 nano leads
GPT-5.4 nano: 43.6 (#84), Qwen-14B: 31.2 (#288)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Coding | 1405 | 1071 |
| SciCode | 46.9% | — |
| WeirdML | 49.2% | — |
| ALE-Bench | 1,005 | — |
Reasoning GPT-5.4 nano leads
GPT-5.4 nano: 23.7 (#173), Qwen-14B: 19.6 (#257)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Hard Prompts | 1381 | 1027 |
| Epoch Capabilities Index | 145.81 | 113.03 |
| ARC-AGI-2 | 5.7% | — |
| Kagi LLM Benchmark | 39.7% | — |
| ARC-AGI-1 | 51.5% | — |
| CritPt | 9.3% | — |
| Chess Puzzles | 30% | — |
| Mystery Game Puzzles | 9% | — |
| DTBench | 80.3% | — |
| LMCA | 36.9% | — |
| BIG-Bench Hard | — | 55% |
| ForecastBench | 57.3 | — |
| LAMBADA | — | 71.1% |
| PIQA | — | 79.9% |
Math GPT-5.4 nano leads
GPT-5.4 nano: 40.9 (#88), Qwen-14B: 31.2 (#227)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Math | 1406 | 1068 |
| FrontierMath (Tiers 1-3) | 44.9% | — |
| FrontierMath Tier 4 | 12.2% | — |
| OTIS Mock AIME 2024-2025 | 87.8% | — |
| ProofBench | 5% | — |
| FrontierMath (Feb 2025 set) | 25.9% | — |
| FrontierMath Tier 4 (v1) | 6.3% | — |
| GSM8K | — | 61.3% |
Knowledge Not comparable
GPT-5.4 nano: 41.9 (#103), Qwen-14B: —
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| GPQA Diamond | 78.5% | — |
| SimpleQA Verified | 11.7% | — |
| Vectara Hallucination Rate | 3.1% | — |
| LMArena Expert | 1396 | — |
| ARC (AI2) Challenge | — | 84.4% |
| BoolQ | — | 86.2% |
| MMLU | — | 66.3% |
Multimodal Not comparable
GPT-5.4 nano: 36.7 (#78), Qwen-14B: —
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Vision | 1196 | — |
Multilingual GPT-5.4 nano leads
GPT-5.4 nano: 48.6 (#140), Qwen-14B: 27.5 (#275)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Non-English | 1359 | 1041 |
| LMArena Chinese | 1392 | 1077 |
| LMArena French | 1396 | — |
| LMArena German | 1367 | — |
| LMArena Japanese | 1343 | — |
| LMArena Korean | 1320 | — |
| LMArena Russian | 1363 | — |
| LMArena Spanish | 1371 | — |
Instruction Following GPT-5.4 nano leads
GPT-5.4 nano: 71.9 (#144), Qwen-14B: 52.4 (#289)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Instruction Following | 1362 | 1031 |
Long Context GPT-5.4 nano leads
GPT-5.4 nano: 41.6 (#137), Qwen-14B: 31.3 (#280)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Longer Query | 1366 | 1028 |
Writing & Preference GPT-5.4 nano leads
GPT-5.4 nano: 55.7 (#142), Qwen-14B: 27.6 (#299)
| Benchmark | GPT-5.4 nano | Qwen-14B |
|---|---|---|
| LMArena Text | 1372 | 1051 |
| LMArena Creative Writing | 1314 | 1028 |
| LMArena Multi-Turn | 1382 | 1022 |
Frequently asked questions
Is GPT-5.4 nano better than Qwen-14B?
GPT-5.4 nano is the stronger model overall, scoring 41.9 to 31.4 on the Noometry Index.
Is GPT-5.4 nano or Qwen-14B better for coding?
GPT-5.4 nano scores higher on coding benchmarks: 43.6 versus 31.2 in the Noometry coding category.
How many benchmarks do GPT-5.4 nano and Qwen-14B share?
11 benchmarks have published results for both models. GPT-5.4 nano has 40 scored results on Noometry and Qwen-14B has 18.