Gemma 4 E2B vs LFM2.5-VL-1.6B
Wins 5 of 7 areas
Coding · Reasoning · Facts · Long documents · Following instructions
Wins 1 of 7 areas
Images and charts
Gemma 4 E2B is the stronger all-rounder.LFM2.5-VL-1.6B is better at images and charts.
Scores updated · 34 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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.
- CodingWriting and fixing software20Gemma 4 E2B2 of 2 tests
- ReasoningHard problems that need careful thinking10Gemma 4 E2B1 of 3 tests · 2 ties
- FactsGetting facts right instead of making them up10Gemma 4 E2B1 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10Gemma 4 E2B1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemma 4 E2B1 of 1 test
- Images and chartsUnderstanding pictures, charts and video13LFM2.5-VL-1.6B3 of 4 tests
- 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 three tests each model wins by the widest margin. Scores are out of 100.
Where Gemma 4 E2B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+63.3points ahead
- Harder college exam questions with imagesMMMU-Pro+18.1points ahead
- Code for real scientific research problemsSciCode+17.9points ahead
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemma 4 E2B
- SciCodeGemma 4 E2B by 17.920.93+17.9
- Terminal-Bench HardGemma 4 E2B by 330+3
ReasoningGemma 4 E2B
- GPQA DiamondGemma 4 E2B by 14.443.328.9+14.4
- Humanity's Last Examtie4.85.1tie
- CritPttie00tie
FactsGemma 4 E2B
- AA-Omniscience · Non-hallucinationGemma 4 E2B by 63.367.64.3+63.3
- AA-Omniscience · Accuracytie6.65.8tie
Images and chartsLFM2.5-VL-1.6B
Following instructionsGemma 4 E2B
- IFBenchGemma 4 E2B by 4.93833.1+4.9
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.
- AA-OmniscienceGemma 4 E2B by 60.8-23.6-84.4+60.8
- ChartQA TestLFM2.5-VL-1.6B by 30.743.273.9+30.7
- MMLU-Pro testGemma 4 E2B by 24.244.119.9+24.2
- MMLU testGemma 4 E2B by 22.869.246.4+22.8
- VLM Judge ScoreGemma 4 E2B by 18.484.466+18.4
- MM-IFEvalGemma 4 E2B by 13.365.652.3+13.3
- τ²-Bench Telecom (AA run)Gemma 4 E2B by 12.320.88.5+12.3
- MMMU DEV_VALGemma 4 E2B by 9.747.738+9.7
- RealWorldQALFM2.5-VL-1.6B by 4.86064.8+4.8
- AA IntelligenceGemma 4 E2B by 37.84.8+3
- MMMBGemma 4 E2B by 2.674.371.7+2.6
- MTL MMBench_DEVGemma 4 E2B by 2.564.862.3+2.5
- DocVQA-valLFM2.5-VL-1.6B by 285.787.7+2
- OCRBench v2_enGemma 4 E2B by 243.541.5+2
- AA Agentic IndexLFM2.5-VL-1.6B by 1.31.52.8+1.3
- GQA TestDev_Balancedtie38.739.5tie
- MMMU (val) (Pass@1)tie41.140.6tie
- HallusionBenchtie59.860.1tie
- MMBench DEV_EN_V11tie69.369.6tie
- MMBench_DEVtie0.60.6tie
Questions people ask
Which is better, Gemma 4 E2B or LFM2.5-VL-1.6B?
Gemma 4 E2B wins five of the seven areas where both have results: coding, reasoning, facts, long documents and following instructions. LFM2.5-VL-1.6B wins images and charts. They are level on agents.
Which is better for coding?
Gemma 4 E2B. It wins 2 of the 2 coding tests both models report; LFM2.5-VL-1.6B wins none.
How do you compare the two?
We use the 34 benchmark tests both models have published scores on. The verdict counts the 14 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.