Phi 3.5 MoE Instruct is the stronger all-rounder.
Scores updated · 14 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.
What it costs
Prices per million tokens, roughly 750,000 words. The bars show the cost of a million tokens read plus a million written.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where Phi 3.5 MoE Instruct pulls ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+7.2points ahead
Where Llama 3 Instruct 8B pulls ahead
No clear win on a test scored out of 100.
Every test, side by side
All 14 tests both models report. The winning score is in its model's colour; marks a score checked independently.
ReasoningPhi 3.5 MoE Instruct
- GPQA DiamondPhi 3.5 MoE Instruct by 7.229.636.8+7.2
Other results13 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.
- GSM8KPhi 3.5 MoE Instruct by 38.849.988.7+38.8
- BBHPhi 3.5 MoE Instruct by 21.457.779.1+21.4
- WinoGrandePhi 3.5 MoE Instruct by 16.36581.3+16.3
- TruthfulQAPhi 3.5 MoE Instruct by 14.363.277.5+14.3
- MBPPPhi 3.5 MoE Instruct by 13.167.780.8+13.1
- OpenBookQAPhi 3.5 MoE Instruct by 1376.689.6+13
- PIQAPhi 3.5 MoE Instruct by 12.975.788.6+12.9
- HellaSwagPhi 3.5 MoE Instruct by 12.771.183.8+12.7
- HumanEvalPhi 3.5 MoE Instruct by 10.360.470.7+10.3
- MMLU-ProPhi 3.5 MoE Instruct by 9.935.445.3+9.9
- ARC-ChallengePhi 3.5 MoE Instruct by 8.282.891+8.2
- Social IQAPhi 3.5 MoE Instruct by 4.173.978+4.1
- BoolQPhi 3.5 MoE Instruct by 3.780.984.6+3.7
Questions people ask
Which is better, Llama 3 Instruct 8B or Phi 3.5 MoE Instruct?
Phi 3.5 MoE Instruct wins the one area where both have results: reasoning. Llama 3 Instruct 8B wins none.
How do you compare the two?
We use the 14 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 13 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.