Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
Qwen3 235B 2507 is clearly ahead on the aggregate, 37 to 33. The gap is large enough that you do not need to squint at the spreadsheet to see the difference.
Qwen3 235B 2507's sharpest advantage is in multimodal & grounded, where it averages 41.6 against 32.4. The single biggest benchmark swing on the page is HumanEval, 31 to 17. LFM2.5-1.2B-Thinking does hit back in instruction following, so the answer changes if that is the part of the workload you care about most.
LFM2.5-1.2B-Thinking is the reasoning model in the pair, while Qwen3 235B 2507 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. Qwen3 235B 2507 gives you the larger context window at 128K, compared with 32K for LFM2.5-1.2B-Thinking.
Pick Qwen3 235B 2507 if you want the stronger benchmark profile. LFM2.5-1.2B-Thinking only becomes the better choice if instruction following is the priority or you want the stronger reasoning-first profile.
Qwen3 235B 2507
33.7
LFM2.5-1.2B-Thinking
34.1
Qwen3 235B 2507
14.6
LFM2.5-1.2B-Thinking
8.2
Qwen3 235B 2507
41.6
LFM2.5-1.2B-Thinking
32.4
Qwen3 235B 2507
45.6
LFM2.5-1.2B-Thinking
38.4
Qwen3 235B 2507
31.9
LFM2.5-1.2B-Thinking
27
Qwen3 235B 2507
69
LFM2.5-1.2B-Thinking
72
Qwen3 235B 2507
60.4
LFM2.5-1.2B-Thinking
60.7
Qwen3 235B 2507
46.6
LFM2.5-1.2B-Thinking
42.3
Qwen3 235B 2507 is ahead overall, 37 to 33. The biggest single separator in this matchup is HumanEval, where the scores are 31 and 17.
Qwen3 235B 2507 has the edge for knowledge tasks in this comparison, averaging 31.9 versus 27. Inside this category, MMLU is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 has the edge for coding in this comparison, averaging 14.6 versus 8.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 has the edge for math in this comparison, averaging 46.6 versus 42.3. Inside this category, AIME 2023 is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 has the edge for reasoning in this comparison, averaging 45.6 versus 38.4. Inside this category, LongBench v2 is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for agentic tasks in this comparison, averaging 34.1 versus 33.7. Inside this category, BrowseComp is the benchmark that creates the most daylight between them.
Qwen3 235B 2507 has the edge for multimodal and grounded tasks in this comparison, averaging 41.6 versus 32.4. Inside this category, MMMU-Pro is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for instruction following in this comparison, averaging 72 versus 69. Inside this category, IFEval is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for multilingual tasks in this comparison, averaging 60.7 versus 60.4. Inside this category, MGSM 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.