DeepSeek-R1-Distill-Llama-70B vs DeepSeek-V3
Wins 0 of 5 areas
—
Wins 4 of 5 areas
Coding · Reasoning · Long documents · Following instructions
DeepSeek-V3 is the stronger all-rounder.
Scores updated · 19 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 software02DeepSeek-V32 of 2 tests
- ReasoningHard problems that need careful thinking01DeepSeek-V31 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts01DeepSeek-V31 of 1 test
- Following instructionsDoing exactly what it is asked01DeepSeek-V31 of 1 test
- FactsGetting facts right instead of making them up11Even1 each
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.
DeepSeek-V3 costs 37% 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-V3 pulls ahead
- Reasons across sets of long documentsAA-LCR+30.7points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+25.3points ahead
- Hard command-line tasks in a real terminalTerminal-Bench Hard+13.7points ahead
Where DeepSeek-R1-Distill-Llama-70B pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+4.8points ahead
Every test, side by side
All 19 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V3
- Terminal-Bench HardDeepSeek-V3 by 13.71.515.2+13.7
- SciCodeDeepSeek-V3 by 7.731.339+7.7
ReasoningDeepSeek-V3
- GPQA DiamondDeepSeek-V3 by 25.340.265.5+25.3
- Humanity's Last Examtie5.14.7tie
- CritPttie00tie
FactsEven
- AA-Omniscience · AccuracyDeepSeek-V3 by 6.319.225.4+6.3
- AA-Omniscience · Non-hallucinationDeepSeek-R1-Distill-Llama-70B by 4.818.914.1+4.8
Following instructionsDeepSeek-V3
- IFBenchDeepSeek-V3 by 13.427.641+13.4
Other results10 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.
- Codeforces (Rating)DeepSeek-R1-Distill-Llama-70B by 499 rating points16331134+499 rating
- AIME 2024DeepSeek-R1-Distill-Llama-70B by 30.87039.2+30.8
- τ²-Bench Telecom (AA run)DeepSeek-V3 by 25.221.947.1+25.2
- Artificial Analysis Coding IndexDeepSeek-V3 by 11.611.423+11.6
- LiveCodeBenchDeepSeek-R1-Distill-Llama-70B by 8.357.549.2+8.3
- AA-OmniscienceDeepSeek-V3 by 5.7-46.4-40.7+5.7
- MATH-500 (EM)DeepSeek-R1-Distill-Llama-70B by 4.394.590.2+4.3
- HMMT Feb. 2025DeepSeek-R1-Distill-Llama-70B by 4.133.329.2+4.1
- AIME 2025DeepSeek-R1-Distill-Llama-70B by 3.75551.3+3.7
- AA IntelligenceDeepSeek-V3 by 1.87.99.7+1.8
Questions people ask
Which is better, DeepSeek-R1-Distill-Llama-70B or DeepSeek-V3?
DeepSeek-V3 wins four of the five areas where both have results: coding, reasoning, long documents and following instructions. DeepSeek-R1-Distill-Llama-70B wins none. They are level on facts.
Which is better for coding?
DeepSeek-V3. It wins 2 of the 2 coding tests both models report; DeepSeek-R1-Distill-Llama-70B wins none.
Which is cheaper?
DeepSeek-R1-Distill-Llama-70B costs $0.70 per million input tokens and $1.10 per million output tokens; DeepSeek-V3 costs $0.24 and $0.90. That makes DeepSeek-V3 about 37% cheaper for the same work.
How do you compare the two?
We use the 19 benchmark tests both models have published scores on. The verdict counts the 9 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 10 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.