Side-by-side benchmark comparison across agentic, coding, multimodal, knowledge, reasoning, and math workflows.
DBRX Instruct 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.
DBRX Instruct's sharpest advantage is in coding, where it averages 48 against 8.2. The single biggest benchmark swing on the page is HumanEval, 70.1 to 17. LFM2.5-1.2B-Thinking does hit back in multilingual, 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 DBRX Instruct 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.
Pick DBRX Instruct if you want the stronger benchmark profile. LFM2.5-1.2B-Thinking only becomes the better choice if multilingual is the priority or you want the stronger reasoning-first profile.
DBRX Instruct
34.3
LFM2.5-1.2B-Thinking
34.1
DBRX Instruct
48
LFM2.5-1.2B-Thinking
8.2
DBRX Instruct
35.6
LFM2.5-1.2B-Thinking
32.4
DBRX Instruct
36.4
LFM2.5-1.2B-Thinking
38.4
DBRX Instruct
52
LFM2.5-1.2B-Thinking
27
Comparable scores for this category are coming soon. One or both models do not have sourced results here yet.
DBRX Instruct
46
LFM2.5-1.2B-Thinking
60.7
Comparable scores for this category are coming soon. One or both models do not have sourced results here yet.
DBRX Instruct is ahead overall, 37 to 33. The biggest single separator in this matchup is HumanEval, where the scores are 70.1 and 17.
DBRX Instruct has the edge for knowledge tasks in this comparison, averaging 52 versus 27. Inside this category, MMLU is the benchmark that creates the most daylight between them.
DBRX Instruct has the edge for coding in this comparison, averaging 48 versus 8.2. Inside this category, HumanEval is the benchmark that creates the most daylight between them.
LFM2.5-1.2B-Thinking has the edge for reasoning in this comparison, averaging 38.4 versus 36.4. Inside this category, MRCRv2 is the benchmark that creates the most daylight between them.
DBRX Instruct has the edge for agentic tasks in this comparison, averaging 34.3 versus 34.1. Inside this category, Terminal-Bench 2.0 is the benchmark that creates the most daylight between them.
DBRX Instruct has the edge for multimodal and grounded tasks in this comparison, averaging 35.6 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 multilingual tasks in this comparison, averaging 60.7 versus 46. Inside this category, MMLU-ProX 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.