LFM2.5-350M vs Llama 4 Behemoth

Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.

Agentic
Coding
Multimodal & Grounded
Reasoning
Knowledge
Instruction Following
Multilingual
Mathematics

LFM2.5-350M· Llama 4 Behemoth

Quick Verdict

Pick LFM2.5-350M if you want the stronger benchmark profile. Llama 4 Behemoth only becomes the better choice if knowledge is the priority.

LFM2.5-350M is clearly ahead on the aggregate, 39 to 35. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.

LFM2.5-350M's sharpest advantage is in instruction following, where it averages 77 against 68. The single biggest benchmark swing on the page is MMLU-Pro, 20.0% to 54%. Llama 4 Behemoth does hit back in knowledge, so the answer changes if that is the part of the workload you care about most.

Operational tradeoffs

ProviderLiquidAIMeta
PriceFree*Free*
SpeedN/AN/A
TTFTN/AN/A
Context32K32K

Decision framing

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.

BenchmarkLFM2.5-350MLlama 4 Behemoth
Agentic
Terminal-Bench 2.033%
BrowseComp38%
OSWorld-Verified34%
Coding
HumanEval40%
SWE-bench Verified15%
LiveCodeBench13%
SWE-bench Pro15%
Multimodal & Grounded
MMMU-Pro60%
OfficeQA Pro49%
Reasoning
MuSR44%
BBH62%
LongBench v246%
MRCRv246%
KnowledgeLlama 4 Behemoth wins
GPQA30.6%47%
MMLU-Pro20.0%54%
MMLU48%
SuperGPQA45%
HLE3%
FrontierScience43%
SimpleQA46%
Instruction FollowingLFM2.5-350M wins
IFEval77.0%68%
Multilingual
MGSM66%
MMLU-ProX61%
Mathematics
AIME 202348%
AIME 202450%
AIME 202549%
HMMT Feb 202344%
HMMT Feb 202446%
HMMT Feb 202545%
BRUMO 202547%
MATH-50060%
Frequently Asked Questions (3)

Which is better, LFM2.5-350M or Llama 4 Behemoth?

LFM2.5-350M is ahead overall, 39 to 35. The biggest single separator in this matchup is MMLU-Pro, where the scores are 20.0% and 54%.

Which is better for knowledge tasks, LFM2.5-350M or Llama 4 Behemoth?

Llama 4 Behemoth has the edge for knowledge tasks in this comparison, averaging 37.3 versus 23.8. Inside this category, MMLU-Pro is the benchmark that creates the most daylight between them.

Which is better for instruction following, LFM2.5-350M or Llama 4 Behemoth?

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.

Last updated: March 31, 2026

Weekly LLM Benchmark Digest

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.