DeepSeek-R1 vs Llama 3.1 Nemotron Ultra 253B V1
Wins 2 of 6 areas
Coding · Long documents
Wins 0 of 6 areas
—
DeepSeek-R1 wins more areas, narrowly.
Scores updated · 18 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.
- CodingWriting and fixing software20DeepSeek-R12 of 2 tests
- Long documentsFinding answers in very long texts10DeepSeek-R11 of 1 test
- AgentsCarrying out multi-step tasks on its own00Even0 each · 1 tie
- ReasoningHard problems that need careful thinking11Even1 each · 1 tie
- FactsGetting facts right instead of making them up11Even1 each
- Following instructionsDoing exactly what it is asked00Even0 each · 1 tie
Agents, 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 DeepSeek-R1 pulls ahead
- Reasons across sets of long documentsAA-LCR+50points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+10.4points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+3.8points ahead
Where Llama 3.1 Nemotron Ultra 253B V1 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.1points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+2points ahead
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.
CodingDeepSeek-R1
- Terminal-Bench HardDeepSeek-R1 by 3.86.12.3+3.8
- SciCodeDeepSeek-R1 by 3.638.334.7+3.6
ReasoningEven
- GPQA DiamondLlama 3.1 Nemotron Ultra 253B V1 by 270.872.8+2
- Humanity's Last ExamDeepSeek-R1 by 1.18.57.4+1.1
- CritPttie0.60tie
FactsEven
- AA-Omniscience · AccuracyDeepSeek-R1 by 10.430.520.1+10.4
- AA-Omniscience · Non-hallucinationLlama 3.1 Nemotron Ultra 253B V1 by 9.19.718.8+9.1
Other results8 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.
- AA-OmniscienceDeepSeek-R1 by 12.7-32.2-44.9+12.7
- Artificial Analysis Coding IndexDeepSeek-R1 by 11.524.613.1+11.5
- IFEvalLlama 3.1 Nemotron Ultra 253B V1 by 6.283.389.5+6.2
- AA IntelligenceDeepSeek-R1 by 3.911.47.5+3.9
- LiveCodeBenchLlama 3.1 Nemotron Ultra 253B V1 by 2.863.566.3+2.8
- AIME 2025Llama 3.1 Nemotron Ultra 253B V1 by 2.57072.5+2.5
- MATH-500 (EM)tie97.397tie
- τ²-Bench Telecom (AA run)tie11.411.4tie
Questions people ask
Which is better, DeepSeek-R1 or Llama 3.1 Nemotron Ultra 253B V1?
DeepSeek-R1 wins two of the six areas where both have results: coding and long documents. Llama 3.1 Nemotron Ultra 253B V1 wins none. They are level on agents, reasoning, facts and following instructions.
Which is better for coding?
DeepSeek-R1. It wins 2 of the 2 coding tests both models report; Llama 3.1 Nemotron Ultra 253B V1 wins none.
How do you compare the two?
We use the 18 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 8 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.