Granite 4.0 1B vs LFM2.5-1.2B-Thinking
Wins 2 of 6 areas
Coding · Long documents
Wins 3 of 6 areas
Reasoning · Facts · Following instructions
LFM2.5-1.2B-Thinking wins more areas, narrowly.Granite 4.0 1B is better at coding.
Scores updated · 20 tests both models report · How we compare
Where each one wins
Tests won in each of the six areas where both have results. Each piece is one test, so a longer bar means more evidence; grey means the two scored within a point of each other.
- ReasoningHard problems that need careful thinking02LFM2.5-1.2B-Thinking2 of 3 tests · 1 tie
- FactsGetting facts right instead of making them up01LFM2.5-1.2B-Thinking1 of 2 tests · 1 tie
- Following instructionsDoing exactly what it is asked01LFM2.5-1.2B-Thinking1 of 1 test
- CodingWriting and fixing software10Granite 4.0 1B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Granite 4.0 1B1 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
Agents, long documents and following instructions rest on a single test each.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where LFM2.5-1.2B-Thinking pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+21.3points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+5.8points ahead
- Very hard expert questions across many subjectsHumanity's Last Exam+1.4points ahead
Where Granite 4.0 1B pulls ahead
- Code for real scientific research problemsSciCode+4.5points ahead
Every test, side by side
All 20 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGranite 4.0 1B
- SciCodeGranite 4.0 1B by 4.58.74.2+4.5
- Terminal-Bench Hardtie00tie
ReasoningLFM2.5-1.2B-Thinking
- GPQA DiamondLFM2.5-1.2B-Thinking by 5.828.133.9+5.8
- Humanity's Last ExamLFM2.5-1.2B-Thinking by 1.44.86.2+1.4
- CritPttie00tie
FactsLFM2.5-1.2B-Thinking
- AA-Omniscience · AccuracyLFM2.5-1.2B-Thinking by 1.26.27.4+1.2
- AA-Omniscience · Non-hallucinationtie6.55.6tie
Following instructionsLFM2.5-1.2B-Thinking
- IFBenchLFM2.5-1.2B-Thinking by 21.320.541.8+21.3
Other results10 tests, not counted
Tests outside the eight areas. They are not counted above: several are summary scores built from other tests, or the same test under another name.
- MATH-500 (EM)LFM2.5-1.2B-Thinking by 43.244.888+43.2
- AIME25 no toolsLFM2.5-1.2B-Thinking by 28.43.331.7+28.4
- MMLU-ProLFM2.5-1.2B-Thinking by 16.233.549.6+16.2
- GSM8KLFM2.5-1.2B-Thinking by 12.273.485.6+12.2
- IFEvalLFM2.5-1.2B-Thinking by 8.879.688.4+8.8
- BFCL v3LFM2.5-1.2B-Thinking by 4.652.457+4.6
- τ²-Bench Telecom (AA run)Granite 4.0 1B by 3.222.819.6+3.2
- AA-OmniscienceLFM2.5-1.2B-Thinking by 1.7-81.6-80+1.7
- AA Agentic IndexGranite 4.0 1B by 1.17.66.5+1.1
- AA Intelligencetie55.2tie
Questions people ask
Which is better, Granite 4.0 1B or LFM2.5-1.2B-Thinking?
LFM2.5-1.2B-Thinking wins three of the six areas where both have results: reasoning, facts and following instructions. Granite 4.0 1B wins coding and long documents. They are level on agents.
Which is better for coding?
Granite 4.0 1B. It wins 1 of the 2 coding tests both models report; LFM2.5-1.2B-Thinking wins none, and 1 is a tie.
How do you compare the two?
We use the 20 benchmark tests both models have published scores on. The verdict counts the 10 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 are listed but not counted, because several are summary scores or repeat a test. Each score is the one shown on the model's own page.