Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
1-bit Bonsai 4B
~44
1/8 categoriesGPT-5.4 mini
68
Winner · 3/8 categories1-bit Bonsai 4B· GPT-5.4 mini
Pick GPT-5.4 mini if you want the stronger benchmark profile. 1-bit Bonsai 4B only becomes the better choice if reasoning is the priority or you want the cheaper token bill.
GPT-5.4 mini is clearly ahead on the aggregate, 68 to 44. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
GPT-5.4 mini's sharpest advantage is in mathematics, where it averages 97.4 against 65.8. The single biggest benchmark swing on the page is GPQA, 28.7% to 88%. 1-bit Bonsai 4B does hit back in reasoning, so the answer changes if that is the part of the workload you care about most.
GPT-5.4 mini is also the more expensive model on tokens at $0.75 input / $4.50 output per 1M tokens, versus $0.00 input / $0.00 output per 1M tokens for 1-bit Bonsai 4B. That is roughly Infinityx on output cost alone. GPT-5.4 mini is the reasoning model in the pair, while 1-bit Bonsai 4B is not. That usually helps on harder chain-of-thought-heavy tests, but it can also mean more latency and more token spend in real use. GPT-5.4 mini gives you the larger context window at 400K, compared with 32K for 1-bit Bonsai 4B.
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 4B | GPT-5.4 mini |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 60% |
| OSWorld-Verified | — | 72.1% |
| MCP Atlas | — | 57.7% |
| Toolathlon | — | 42.9% |
| tau2-bench | — | 93.4% |
| Coding | ||
| SWE-bench Pro | — | 54.4% |
| Multimodal & Grounded | ||
| MMMU-Pro | — | 76.6% |
| MMMU-Pro w/ Python | — | 78% |
| OmniDocBench 1.5 | — | 0.1263 |
| Reasoning1-bit Bonsai 4B wins | ||
| MuSR | 41.4% | — |
| MRCRv2 | — | 40.7% |
| MRCR v2 64K-128K | — | 47.7% |
| MRCR v2 128K-256K | — | 33.6% |
| Graphwalks BFS 128K | — | 76.3% |
| Graphwalks Parents 128K | — | 71.5% |
| KnowledgeGPT-5.4 mini wins | ||
| GPQA | 28.7% | 88% |
| HLE | — | 41.5% |
| HLE w/o tools | — | 28.2% |
| Instruction FollowingGPT-5.4 mini wins | ||
| IFEval | 69.6% | 87.4% |
| Multilingual | ||
| Coming soon | ||
| MathematicsGPT-5.4 mini wins | ||
| MATH-500 | 65.8% | 97.4% |
GPT-5.4 mini is ahead overall, 68 to 44. The biggest single separator in this matchup is GPQA, where the scores are 28.7% and 88%.
GPT-5.4 mini has the edge for knowledge tasks in this comparison, averaging 57.4 versus 28.7. Inside this category, GPQA is the benchmark that creates the most daylight between them.
GPT-5.4 mini has the edge for math in this comparison, averaging 97.4 versus 65.8. Inside this category, MATH-500 is the benchmark that creates the most daylight between them.
1-bit Bonsai 4B has the edge for reasoning in this comparison, averaging 41.4 versus 40.7. GPT-5.4 mini stays close enough that the answer can still flip depending on your workload.
GPT-5.4 mini has the edge for instruction following in this comparison, averaging 87.4 versus 69.6. 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.