DeepSeek-R1 vs Qwen2.5 Instruct 72B
Wins 4 of 5 areas
Coding · Reasoning · Long documents · Following instructions
Wins 0 of 5 areas
—
DeepSeek-R1 is the stronger all-rounder.Qwen2.5 Instruct 72B is cheaper.
Scores updated · 36 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.
- CodingWriting and fixing software30DeepSeek-R13 of 3 tests
- ReasoningHard problems that need careful thinking20DeepSeek-R12 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts20DeepSeek-R12 of 2 tests
- Following instructionsDoing exactly what it is asked10DeepSeek-R11 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
Following instructions 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.
Qwen2.5 Instruct 72B costs 84% 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
- Reasons across sets of long documentsAA-LCR+37.7points ahead
- Fixes real GitHub issues in Python projectsSWE-bench Verified+25.4points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+21.7points ahead
Where Qwen2.5 Instruct 72B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+4.8points ahead
Every test, side by side
All 36 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-R1
- SWE-bench VerifiedDeepSeek-R1 by 25.449.223.8+25.4
- SciCodeDeepSeek-R1 by 11.638.326.7+11.6
- Terminal-Bench HardDeepSeek-R1 by 1.66.14.5+1.6
ReasoningDeepSeek-R1
- GPQA DiamondDeepSeek-R1 by 21.770.849.1+21.7
- Humanity's Last ExamDeepSeek-R1 by 4.98.53.6+4.9
- CritPttie0.60tie
FactsEven
- AA-Omniscience · AccuracyDeepSeek-R1 by 1330.517.5+13
- AA-Omniscience · Non-hallucinationQwen2.5 Instruct 72B by 4.89.714.5+4.8
Long documentsDeepSeek-R1
- AA-LCRDeepSeek-R1 by 37.757.720+37.7
- LongBench v2DeepSeek-R1 by 18.958.339.4+18.9
Following instructionsDeepSeek-R1
- IFBenchDeepSeek-R1 by 2.13936.9+2.1
Other results25 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.
- CodeforcesDeepSeek-R1 by 2004 rating points202924.8+2004 rating
- AIME 2024DeepSeek-R1 by 56.579.823.3+56.5
- HarmBenchQwen2.5 Instruct 72B by 24.947.972.8+24.9
- τ²-Bench Telecom (AA run)Qwen2.5 Instruct 72B by 23.111.434.5+23.1
- SimpleQADeepSeek-R1 by 2130.19.1+21
- AA-OmniscienceDeepSeek-R1 by 20.9-32.2-53.1+20.9
- MATH-500 (EM)DeepSeek-R1 by 17.397.380+17.3
- DROP (3-shot F1)DeepSeek-R1 by 15.592.276.7+15.5
- Chinese SimpleQA (C-SimpleQA)DeepSeek-R1 by 15.363.748.4+15.3
- MMLUDeepSeek-R1 by 13.890.877+13.8
- MMLU-ProDeepSeek-R1 by 12.98471.1+12.9
- Artificial Analysis Coding IndexDeepSeek-R1 by 12.724.611.9+12.7
- FRAMES (Acc.)DeepSeek-R1 by 12.782.569.8+12.7
- CLUEWSCDeepSeek-R1 by 10.392.882.5+10.3
- MATHDeepSeek-R1 by 8.997.388.4+8.9
- LiveCodeBenchDeepSeek-R1 by 863.555.5+8
- AIR-Bench 2024Qwen2.5 Instruct 72B by 6.152.959+6.1
- MMLU-ReduxDeepSeek-R1 by 6.192.986.8+6.1
- AA IntelligenceDeepSeek-R1 by 3.711.47.7+3.7
- XSTestQwen2.5 Instruct 72B by 3.594.497.9+3.5
- C-EvalDeepSeek-R1 by 2.691.889.2+2.6
- anthropic_red_teamQwen2.5 Instruct 72B by 2.497.299.6+2.4
- simple_safety_testsQwen2.5 Instruct 72B by 298100+2
- bbqDeepSeek-R1 by 1.296.695.4+1.2
- IFEvaltie83.384.1tie
Questions people ask
Which is better, DeepSeek-R1 or Qwen2.5 Instruct 72B?
DeepSeek-R1 wins four of the five areas where both have results: coding, reasoning, long documents and following instructions. Qwen2.5 Instruct 72B wins none, but costs 84% less. They are level on facts.
Which is better for coding?
DeepSeek-R1. It wins 3 of the 3 coding tests both models report; Qwen2.5 Instruct 72B wins none.
Which is cheaper?
DeepSeek-R1 costs $2.00 per million input tokens and $4.00 per million output tokens; Qwen2.5 Instruct 72B costs $0.47 and $0.49. That makes Qwen2.5 Instruct 72B about 84% cheaper for the same work.
How do you compare the two?
We use the 36 benchmark tests both models have published scores on. The verdict counts the 11 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 25 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.