Phi 3 Mini 4K Instruct is the stronger all-rounder.
Scores updated · 18 tests both models report · How we compare
Where each one wins
Tests won in each of the one area 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.
Reasoning 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 Phi 3 Mini 4K Instruct pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+6.2points ahead
Where Gemma (7B) pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 18 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningPhi 3 Mini 4K Instruct
- GPQA DiamondPhi 3 Mini 4K Instruct by 6.225.731.9+6.2
Other results17 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.
- ARC-ChallengePhi 3 Mini 4K Instruct by 25.261.186.3+25.2
- HumanEvalPhi 3 Mini 4K Instruct by 20.137.257.3+20.1
- TruthfulQAPhi 3 Mini 4K Instruct by 19.944.864.7+19.9
- MBPPPhi 3 Mini 4K Instruct by 19.250.669.8+19.2
- MMLUPhi 3 Mini 4K Instruct by 13.857.170.9+13.8
- BoolQPhi 3 Mini 4K Instruct by 12.16678.1+12.1
- Social IQAPhi 3 Mini 4K Instruct by 12.165.577.6+12.1
- TriviaQAGemma (7B) by 10.972.361.4+10.9
- WinoGrandeGemma (7B) by 7.47971.6+7.4
- HellaSwagGemma (7B) by 6.982.275.3+6.9
- MedQAPhi 3 Mini 4K Instruct by 6.949.656.5+6.9
- PIQAPhi 3 Mini 4K Instruct by 5.478.183.5+5.4
- ANLIPhi 3 Mini 4K Instruct by 4.948.753.6+4.9
- AGI EVALGemma (7B) by 3.142.139+3.1
- Popular aggregated benchmarkPhi 3 Mini 4K Instruct by 1.759.461.1+1.7
- OpenBookQAPhi 3 Mini 4K Instruct by 1.480.882.2+1.4
- MATHtie5050.6tie
Questions people ask
Which is better, Gemma (7B) or Phi 3 Mini 4K Instruct?
Phi 3 Mini 4K Instruct wins the one area where both have results: reasoning. Gemma (7B) wins none.
How do you compare the two?
We use the 18 benchmark tests both models have published scores on. The verdict counts the 1 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 17 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.