DeepSeek-R1 vs Phi 4
Wins 4 of 5 areas
Coding · Reasoning · Long documents · Following instructions
Wins 1 of 5 areas
Facts
DeepSeek-R1 is the stronger all-rounder.Phi 4 is cheaper and better at facts.
Scores updated · 23 tests both models report · How we compare
Where each one wins
Tests won in each of the five 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 thinking20DeepSeek-R12 of 3 tests · 1 tie
- CodingWriting and fixing software20DeepSeek-R12 of 2 tests
- Long documentsFinding answers in very long texts10DeepSeek-R11 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-R11 of 1 test
- FactsGetting facts right instead of making them up12Phi 42 of 3 tests
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.
Phi 4 costs 90% less for the same work.
The biggest differences
The tests each model wins by the widest margin, up to three each. Scores are out of 100.
Where DeepSeek-R1 pulls ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+16.4points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+15.5points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+13.3points ahead
Where Phi 4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.1points 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.
CodingDeepSeek-R1
- SciCodeDeepSeek-R1 by 12.338.326+12.3
- Terminal-Bench HardDeepSeek-R1 by 2.36.13.8+2.3
ReasoningDeepSeek-R1
- GPQA DiamondDeepSeek-R1 by 13.370.857.5+13.3
- Humanity's Last ExamDeepSeek-R1 by 4.78.53.8+4.7
- CritPttie0.60tie
FactsPhi 4
- AA-Omniscience · AccuracyDeepSeek-R1 by 16.430.514.1+16.4
- AA-Omniscience · Non-hallucinationPhi 4 by 9.19.718.8+9.1
- Vectara HHEM hallucination ratelower is betterPhi 4 by 7.611.33.7+7.6
Following instructionsDeepSeek-R1
- IFBenchDeepSeek-R1 by 15.53923.5+15.5
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.
- SimpleQADeepSeek-R1 by 27.130.13+27.1
- MMLU-ProXDeepSeek-R1 by 25.675.549.9+25.6
- AA-OmniscienceDeepSeek-R1 by 23.5-32.2-55.7+23.5
- IFEvalDeepSeek-R1 by 20.383.363+20.3
- MATHDeepSeek-R1 by 16.997.380.4+16.9
- vectara_answer_rateDeepSeek-R1 by 16.39780.7+16.3
- MMLU-ProDeepSeek-R1 by 13.68470.4+13.6
- Artificial Analysis Coding IndexDeepSeek-R1 by 13.424.611.2+13.4
- τ²-Bench Telecom (AA run)DeepSeek-R1 by 11.411.40+11.4
- vectara_factual_consistencyPhi 4 by 7.688.796.3+7.6
- MMLUDeepSeek-R1 by 690.884.8+6
- AA IntelligenceDeepSeek-R1 by 5.511.45.9+5.5
- vectara_avg_summary_lengthPhi 4 by 27.4 rating points93.5120.9+27.4 rating
Questions people ask
Which is better, DeepSeek-R1 or Phi 4?
DeepSeek-R1 wins four of the five areas where both have results: coding, reasoning, long documents and following instructions. Phi 4 wins facts, and costs 90% less.
Which is better for coding?
DeepSeek-R1. It wins 2 of the 2 coding tests both models report; Phi 4 wins none.
Which is cheaper?
DeepSeek-R1 costs $2.00 per million input tokens and $4.00 per million output tokens; Phi 4 costs $0.13 and $0.50. That makes Phi 4 about 90% cheaper for the same work.
How do you compare the two?
We use the 23 benchmark tests both models have published scores on. The verdict counts the 10 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.