Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
1-bit Bonsai 1.7B
~39
0/8 categoriesQwen2.5-VL-32B
~50
Winner · 1/8 categories1-bit Bonsai 1.7B· Qwen2.5-VL-32B
Pick Qwen2.5-VL-32B if you want the stronger benchmark profile. 1-bit Bonsai 1.7B only becomes the better choice if its workflow or ecosystem matters more than the raw scoreboard.
Qwen2.5-VL-32B is clearly ahead on the aggregate, 50 to 39. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen2.5-VL-32B's sharpest advantage is in knowledge, where it averages 60.8 against 20.7. The single biggest benchmark swing on the page is GPQA, 20.7% to 46%.
BenchLM keeps the benchmark table and the operator tradeoffs on the same page so a better score does not hide a materially slower, pricier, or smaller-context model.
Runtime metrics show N/A when BenchLM does not have a sourced snapshot for that exact model. The scoring rules and freshness policy are documented on the methodology page.
| Benchmark | 1-bit Bonsai 1.7B | Qwen2.5-VL-32B |
|---|---|---|
| Agentic | ||
| Coming soon | ||
| Coding | ||
| HumanEval | — | 91.5% |
| Multimodal & Grounded | ||
| MMMU-Pro | — | 49.5% |
| Reasoning | ||
| MuSR | 45.1% | — |
| KnowledgeQwen2.5-VL-32B wins | ||
| GPQA | 20.7% | 46% |
| MMLU-Pro | — | 68.8% |
| Instruction Following | ||
| IFEval | 63% | — |
| Multilingual | ||
| Coming soon | ||
| Mathematics | ||
| MATH-500 | 34.4% | — |
Qwen2.5-VL-32B is ahead overall, 50 to 39. The biggest single separator in this matchup is GPQA, where the scores are 20.7% and 46%.
Qwen2.5-VL-32B has the edge for knowledge tasks in this comparison, averaging 60.8 versus 20.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
Get notified when new models drop, benchmark scores change, or the leaderboard shifts. One email per week.
Free. No spam. Unsubscribe anytime. We only store derived location metadata for consent routing.