Gemma 4 26B A4B vs Phi 4 Mini Instruct
Wins 6 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents · Following instructions
Wins 0 of 6 areas
—
Gemma 4 26B A4B is the stronger all-rounder.
Scores updated · 24 tests both models report · How we compare
Where each one wins
Tests won in each of the six 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 thinking20Gemma 4 26B A4B2 of 3 tests · 1 tie
- CodingWriting and fixing software20Gemma 4 26B A4B2 of 2 tests
- AgentsCarrying out multi-step tasks on its own20Gemma 4 26B A4B2 of 2 tests
- FactsGetting facts right instead of making them up21Gemma 4 26B A4B2 of 3 tests
- Long documentsFinding answers in very long texts10Gemma 4 26B A4B1 of 1 test
- Following instructionsDoing exactly what it is asked10Gemma 4 26B A4B1 of 1 test
Long documents and following instructions rest on a single test each.
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 Gemma 4 26B A4B pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+51.3points ahead
- Reasons across sets of long documentsAA-LCR+50.4points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+46.1points ahead
Where Phi 4 Mini Instruct pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+8.4points ahead
Every test, side by side
All 24 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingGemma 4 26B A4B
- SciCodeGemma 4 26B A4B by 29.24010.8+29.2
- Terminal-Bench HardGemma 4 26B A4B by 13.613.60+13.6
AgentsGemma 4 26B A4B
- τ-Bench V3 · BankingGemma 4 26B A4B by 11121+11
- GDPValGemma 4 26B A4B by 3.43.40+3.4
ReasoningGemma 4 26B A4B
- GPQA DiamondGemma 4 26B A4B by 46.179.233.1+46.1
- Humanity's Last ExamGemma 4 26B A4B by 14.819.34.5+14.8
- CritPttie00tie
FactsGemma 4 26B A4B
- Vectara HHEM hallucination ratelower is betterGemma 4 26B A4B by 18.35.223.5+18.3
- AA-Omniscience · AccuracyGemma 4 26B A4B by 9.619.19.5+9.6
- AA-Omniscience · Non-hallucinationPhi 4 Mini Instruct by 8.413.622+8.4
Following instructionsGemma 4 26B A4B
- IFBenchGemma 4 26B A4B by 51.372.421.1+51.3
Other results12 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.
- Artificial Analysis Coding IndexGemma 4 26B A4B by 35.539.33.8+35.5
- τ²-Bench Telecom (AA run)Gemma 4 26B A4B by 35.443.68.2+35.4
- vectara_avg_summary_lengthPhi 4 Mini Instruct by 353.1 rating points67.1420.2+353.1 rating
- OpenBookQAPhi 4 Mini Instruct by 30.448.879.2+30.4
- MMLU-ProGemma 4 26B A4B by 29.882.652.8+29.8
- vectara_factual_consistencyGemma 4 26B A4B by 18.394.876.5+18.3
- HellaSwagGemma 4 26B A4B by 16.285.369.1+16.2
- WinoGrandeGemma 4 26B A4B by 12.179.167+12.1
- AA IntelligenceGemma 4 26B A4B by 10.416.76.3+10.4
- AA-OmniscienceGemma 4 26B A4B by 10.3-50.8-61.1+10.3
- ARC-ChallengeGemma 4 26B A4B by 9.192.883.7+9.1
- vectara_answer_rateGemma 4 26B A4B by 7.399.892.5+7.3
Questions people ask
Which is better, Gemma 4 26B A4B or Phi 4 Mini Instruct?
Gemma 4 26B A4B wins all six areas where both have results: coding, agents, reasoning, facts, long documents and following instructions. Phi 4 Mini Instruct wins none.
Which is better for coding?
Gemma 4 26B A4B. It wins 2 of the 2 coding tests both models report; Phi 4 Mini Instruct wins none.
How do you compare the two?
We use the 24 benchmark tests both models have published scores on. The verdict counts the 12 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 12 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.