Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
LFM2.5-350M
~39
1/8 categoriesLlama 4 Maverick
39
1/8 categoriesLFM2.5-350M· Llama 4 Maverick
Treat this as a split decision. LFM2.5-350M makes more sense if instruction following is the priority; Llama 4 Maverick is the better fit if knowledge is the priority or you need the larger 1M context window.
LFM2.5-350M and Llama 4 Maverick finish on the same overall score, so this is less about a single winner and more about where the edge shows up. The headline says tie; the benchmark table is where the real choice happens.
Llama 4 Maverick gives you the larger context window at 1M, compared with 32K for LFM2.5-350M.
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 | LFM2.5-350M | Llama 4 Maverick |
|---|---|---|
| Agentic | ||
| Terminal-Bench 2.0 | — | 37% |
| BrowseComp | — | 51% |
| OSWorld-Verified | — | 38% |
| Coding | ||
| HumanEval | — | 38% |
| SWE-bench Verified | — | 13% |
| LiveCodeBench | — | 15% |
| SWE-bench Pro | — | 17% |
| Multimodal & Grounded | ||
| MMMU-Pro | — | 59% |
| OfficeQA Pro | — | 54% |
| Reasoning | ||
| MuSR | — | 42% |
| BBH | — | 63% |
| LongBench v2 | — | 63% |
| MRCRv2 | — | 63% |
| KnowledgeLlama 4 Maverick wins | ||
| GPQA | 30.6% | 45% |
| MMLU-Pro | 20.0% | 53% |
| MMLU | — | 46% |
| SuperGPQA | — | 43% |
| HLE | — | 4% |
| FrontierScience | — | 45% |
| SimpleQA | — | 44% |
| Instruction FollowingLFM2.5-350M wins | ||
| IFEval | 77.0% | 68% |
| Multilingual | ||
| MGSM | — | 63% |
| MMLU-ProX | — | 58% |
| Mathematics | ||
| AIME 2023 | — | 46% |
| AIME 2024 | — | 48% |
| AIME 2025 | — | 47% |
| HMMT Feb 2023 | — | 42% |
| HMMT Feb 2024 | — | 44% |
| HMMT Feb 2025 | — | 43% |
| BRUMO 2025 | — | 45% |
| MATH-500 | — | 59% |
LFM2.5-350M and Llama 4 Maverick are tied on overall score, so the right pick depends on which category matters most for your use case.
Llama 4 Maverick has the edge for knowledge tasks in this comparison, averaging 37 versus 23.8. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.
LFM2.5-350M has the edge for instruction following in this comparison, averaging 77 versus 68. 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.