Gemma 3 1B Instruct vs LFM2.5-1.2B-Instruct
Wins 0 of 5 areas
—
Wins 2 of 5 areas
Reasoning · Following instructions
LFM2.5-1.2B-Instruct wins more areas, narrowly.
Scores updated · 28 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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-Instruct2 of 3 tests · 1 tie
- Following instructionsDoing exactly what it is asked01LFM2.5-1.2B-Instruct1 of 1 test
- CodingWriting and fixing software00Even0 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Long documentsFinding answers in very long texts00Even0 each · 1 tie
Coding, 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-Instruct pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+23.9points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+8.9points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+3.2points ahead
Where Gemma 3 1B Instruct pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+2.8points ahead
Every test, side by side
All 28 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningLFM2.5-1.2B-Instruct
- GPQA DiamondLFM2.5-1.2B-Instruct by 8.923.732.6+8.9
- Humanity's Last ExamLFM2.5-1.2B-Instruct by 1.45.36.7+1.4
- CritPttie00tie
FactsEven
- AA-Omniscience · AccuracyLFM2.5-1.2B-Instruct by 3.23.87+3.2
- AA-Omniscience · Non-hallucinationGemma 3 1B Instruct by 2.817.714.9+2.8
Following instructionsLFM2.5-1.2B-Instruct
- IFBenchLFM2.5-1.2B-Instruct by 23.919.943.8+23.9
Other results20 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.
- GSM8KGemma 3 1B Instruct by 58.562.84.3+58.5
- MATH-500 (EM)Gemma 3 1B Instruct by 39.445.25.8+39.4
- HumanEvalGemma 3 1B Instruct by 36.241.55.3+36.2
- J-MATH500LFM2.5-1.2B-Instruct by 34.415.650+34.4
- BFCL v3LFM2.5-1.2B-Instruct by 32.516.649.1+32.5
- MBPPGemma 3 1B Instruct by 29.735.25.5+29.7
- MMLU-ProLFM2.5-1.2B-Instruct by 29.714.744.4+29.7
- J-BFCLv3LFM2.5-1.2B-Instruct by 2917.346.3+29
- Avg.LFM2.5-1.2B-Instruct by 25.524.650.1+25.5
- JMMLU-ProXLFM2.5-1.2B-Instruct by 17.314.131.4+17.3
- J-GSM8KLFM2.5-1.2B-Instruct by 16.633.650.2+16.6
- Domain AvgLFM2.5-1.2B-Instruct by 15.823.939.7+15.8
- JMMLULFM2.5-1.2B-Instruct by 13.134.547.6+13.1
- JGPQALFM2.5-1.2B-Instruct by 7.524.231.7+7.5
- IFEvalLFM2.5-1.2B-Instruct by 680.286.2+6
- JHumanEval+LFM2.5-1.2B-Instruct by 3.72528.7+3.7
- AA-OmniscienceLFM2.5-1.2B-Instruct by 3.3-75.5-72.1+3.3
- AA Intelligencetie4.85.2tie
- τ²-Bench Telecom (AA run)tie10.510.8tie
- AA Agentic Indextie3.53.6tie
Questions people ask
Which is better, Gemma 3 1B Instruct or LFM2.5-1.2B-Instruct?
LFM2.5-1.2B-Instruct wins two of the five areas where both have results: reasoning and following instructions. Gemma 3 1B Instruct wins none. They are level on coding, facts and long documents.
Which is better for coding?
Neither. All 1 coding tests both models report are ties.
How do you compare the two?
We use the 28 benchmark tests both models have published scores on. The verdict counts the 8 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.