Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
1-bit Bonsai 1.7B
~39
0/8 categoriesGPT-4.1 nano
44
Winner · 3/8 categories1-bit Bonsai 1.7B· GPT-4.1 nano
Pick GPT-4.1 nano if you want the stronger benchmark profile. 1-bit Bonsai 1.7B only becomes the better choice if you want the cheaper token bill.
GPT-4.1 nano is clearly ahead on the aggregate, 44 to 39. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-4.1 nano's sharpest advantage is in knowledge, where it averages 50.7 against 20.7. The single biggest benchmark swing on the page is GPQA, 20.7% to 50.3%.
GPT-4.1 nano is also the more expensive model on tokens at $0.10 input / $0.40 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for 1-bit Bonsai 1.7B. That is roughly Infinityx on output cost alone. GPT-4.1 nano gives you the larger context window at 1M, compared with 32K for 1-bit Bonsai 1.7B.
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 | GPT-4.1 nano |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 43% |
| BrowseComp | — | 62% |
| OSWorld-Verified | — | 42% |
| Coding | ||
| SWE-bench Pro | — | 18% |
| Multimodal & Grounded | ||
| MMMU-Pro | — | 53% |
| OfficeQA Pro | — | 67% |
| ReasoningGPT-4.1 nano wins | ||
| MuSR | 45.1% | — |
| LongBench v2 | — | 75% |
| MRCRv2 | — | 73% |
| KnowledgeGPT-4.1 nano wins | ||
| GPQA | 20.7% | 50.3% |
| MMLU | — | 80.1% |
| FrontierScience | — | 51% |
| Instruction FollowingGPT-4.1 nano wins | ||
| IFEval | 63% | 83.2% |
| Multilingual | ||
| MMLU-ProX | — | 59% |
| Mathematics | ||
| MATH-500 | 34.4% | — |
| AIME 2024 | — | 9.8% |
GPT-4.1 nano is ahead overall, 44 to 39. The biggest single separator in this matchup is GPQA, where the scores are 20.7% and 50.3%.
GPT-4.1 nano has the edge for knowledge tasks in this comparison, averaging 50.7 versus 20.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
GPT-4.1 nano has the edge for reasoning in this comparison, averaging 74.1 versus 45.1. 1-bit Bonsai 1.7B stays close enough that the answer can still flip depending on your workload.
GPT-4.1 nano has the edge for instruction following in this comparison, averaging 83.2 versus 63. Inside this category, IFEval 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.