Up to four models
GPT-5.4 Nano vs Qwen 3.5 Plus (hosted 397B-A17B)
On the index (GPT-4 = 100), GPT-5.4 Nano scores 141 and Qwen 3.5 Plus (hosted 397B-A17B) 141. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.
GPT-5.4 Nano
141125–157
WeightsClosed
ReleasedMar 2026
Price in / out$0.20 / $1.25
Context400K
Core tests taken4 of 9
141124–158
WeightsClosed
ReleasedFeb 2026
Price in / out–
Context–
Core tests taken1 of 9
GPT-5.4 Nano and Qwen 3.5 Plus (hosted 397B-A17B) are too close to call (0 points apart, within the uncertainty), ahead on 0 of 1 shared core tests.
Test by testthird-party results
GPT-5.4 Nano
Qwen 3.5 Plus (hosted 397B-A17B)
Every result as a table
| Test | GPT-5.4 Nano | Qwen 3.5 Plus (hosted 397B-A17B) |
|---|---|---|
| Reasoning & knowledge | ||
| GPQA Diamond (Epoch AI) | 78.5% | 84.9% |
| GPQA Diamond (Vals AI) | 77.5% | 87.4% |
| MMLU Pro | 77.2% | 87.2% |
| MMMU Pro | 73.6% | 22.8% |
| SimpleQA Verified | 11.7% | 25.4% |
| Maths | ||
| AIME-style maths (OTIS mock) | 87.8% | 86.7% |
| Chess puzzles | 45.0% | 9.0% |
| FrontierMath (2025 set) | 25.9% | 21.0% |
| FrontierMath tier 4 (2025 set) | 6.3% | 2.1% |
| Coding | ||
| LiveCodeBench | 84.0% | 85.3% |
| SWE-bench | 69.8% | 71.2% |
| Vibe Code Bench v1.1 | 26.1% | 15.7% |
| Agents | ||
| DTBench | 80.3% | 80.5% |
| LMCA | 36.9% | 36.4% |
| Mystery game puzzles | 9.0% | 17.0% |
| Professional work | ||
| CorpFin v2 | 61.2% | 65.3% |
| LegalBench | 77.9% | 85.1% |
| MortgageTax | 59.1% | 60.8% |
Tests at least two of these models have taken. Bold rows are the core tests.