Compare / Qwen3 30B A3B Instruct 2507 vs GPT-5 Nano
Qwen3 30B A3B Instruct 2507 vs GPT-5 Nano
21 results from 3 sources that measured both models. Each row is in the source's own units; there is no overall winner.
| Benchmark | Metric | Qwen3 30B A3B Instruct 2507 | GPT-5 Nano |
|---|---|---|---|
| Arena (formerly LMArena)reported | Business, management and financeArena rating | 1,397 | 1,339 |
| Arena (formerly LMArena)reported | Creative writingArena rating | 1,319 | 1,247 |
| Arena (formerly LMArena)reported | Expert promptsArena rating | 1,395 | 1,354 |
| Arena (formerly LMArena)reported | Instruction followingArena rating | 1,366 | 1,327 |
| Arena (formerly LMArena)reported | OverallArena rating | 1,383 | 1,338 |
| Arena (formerly LMArena)reported | Writing, literature and languageArena rating | 1,342 | 1,289 |
| Artificial Analysisreported | Artificial Analysis Intelligence Indexindex score | 7.50 | 13.0 |
| Artificial Analysisreported | GPQA Diamond% of questions | 65.9% | 67.6% |
| Artificial Analysisreported | Humanity's Last Exam% of questions | 6.9% | 9.5% |
| Artificial Analysisreported | IFBench% of instructions | 33.1% | 67.6% |
| Artificial Analysisreported | Long-context reasoning (AA-LCR)% of questions | 26.3% | 45.0% |
| Artificial Analysisreported | Terminal-Bench Hard% of tasks | 6.1% | 17.4% |
| Artificial Analysisreported | Τ²-bench telecom% of tasks | 10.2% | 36.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Irrelevance detection% correct | 79.9% | 89.1% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Memory% correct | 17.6% | 24.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Multi-turn tasks% correct | 30.0% | 34.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Overall accuracy% correct | 41.4% | 51.5% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Relevance detection% correct | 93.8% | 93.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (curated)% correct | 88.9% | 80.8% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Single-turn calls (user-contributed)% correct | 78.4% | 70.7% |
| Berkeley Function Calling Leaderboard (BFCL) V4reported | Web search% correct | 22.5% | 72.5% |
Bold green marks the better value on that metric. Where a source reports ranges that overlap, the difference may not be meaningful; see the benchmark page for ranges.