The two are evenly matched.LFM2-8B-A1B (Non-Reasoning) is better at facts; Qwen3 1.7B at reasoning.
Scores updated · 23 tests both models report · How we compare
Where each one wins
Tests won in each of the four 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.
- FactsGetting facts right instead of making them up10LFM2-8B-A1B (Non-Reasoning)1 of 2 tests · 1 tie
- ReasoningHard problems that need careful thinking01Qwen3 1.7B1 of 3 tests · 2 ties
- CodingWriting and fixing software11Even1 each · 1 tie
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Following instructions rests on a single test.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where LFM2-8B-A1B (Non-Reasoning) pulls ahead
- Code for real scientific research problemsSciCode+2.5points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+2.5points ahead
Where Qwen3 1.7B pulls ahead
- Recent programming contest problemsLiveCodeBench v6+3.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+1.2points ahead
Every test, side by side
All 23 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingEven
- LiveCodeBench v6Qwen3 1.7B by 3.12124.1+3.1
- SciCodeLFM2-8B-A1B (Non-Reasoning) by 2.56.84.3+2.5
- Terminal-Bench Hardtie00tie
ReasoningQwen3 1.7B
- GPQA DiamondQwen3 1.7B by 1.234.435.6+1.2
- Humanity's Last Examtie4.94.6tie
- CritPttie00tie
FactsLFM2-8B-A1B (Non-Reasoning)
- AA-Omniscience · Non-hallucinationLFM2-8B-A1B (Non-Reasoning) by 2.57.75.2+2.5
- AA-Omniscience · Accuracytie8.28.9tie
Other results14 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.
- MMLU-ProQwen3 1.7B by 19.637.457+19.6
- AIME25 no toolsQwen3 1.7B by 16.32036.3+16.3
- τ²-Bench Telecom (AA run)Qwen3 1.7B by 15.510.526+15.5
- Creative Writing v3LFM2-8B-A1B (Non-Reasoning) by 12.644.231.6+12.6
- BFCL v3Qwen3 1.7B by 10.345.155.4+10.3
- MMMLULFM2-8B-A1B (Non-Reasoning) by 8.855.346.5+8.8
- HumanEval+LFM2-8B-A1B (Non-Reasoning) by 8.569.561+8.5
- MATH-500 (EM)Qwen3 1.7B by 7.774.281.9+7.7
- MGSMLFM2-8B-A1B (Non-Reasoning) by 5.872.466.6+5.8
- MMLULFM2-8B-A1B (Non-Reasoning) by 5.764.859.1+5.7
- LCB v5Qwen3 1.7B by 5.121.426.5+5.1
- IFEvalLFM2-8B-A1B (Non-Reasoning) by 3.677.674+3.6
- AA-OmniscienceLFM2-8B-A1B (Non-Reasoning) by 1-76.5-77.5+1
- AA Intelligencetie4.95.2tie
Questions people ask
Which is better, LFM2-8B-A1B (Non-Reasoning) or Qwen3 1.7B?
LFM2-8B-A1B (Non-Reasoning) and Qwen3 1.7B each win one of the four areas where both have results. LFM2-8B-A1B (Non-Reasoning) wins facts; Qwen3 1.7B wins reasoning. They are level on coding and following instructions.
Which is better for coding?
Neither. They win 1 coding test each of the 3 both models report, and 1 is a tie.
How do you compare the two?
We use the 23 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 14 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.