DeepSeek-V3 vs Llama 3.3 70B Instruct
Wins 5 of 6 areas
Coding · Agents · Reasoning · Facts · Long documents
Wins 1 of 6 areas
Following instructions
DeepSeek-V3 is the stronger all-rounder.Llama 3.3 70B Instruct is better at following instructions.
Scores updated · 34 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 software30DeepSeek-V33 of 3 tests
- ReasoningHard problems that need careful thinking21DeepSeek-V32 of 4 tests · 1 tie
- FactsGetting facts right instead of making them up21DeepSeek-V32 of 3 tests
- AgentsCarrying out multi-step tasks on its own10DeepSeek-V31 of 2 tests · 1 tie
- Long documentsFinding answers in very long texts10DeepSeek-V31 of 1 test
- Following instructionsDoing exactly what it is asked01Llama 3.3 70B Instruct1 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.
DeepSeek-V3 costs 20% 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+25points ahead
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+15.7points ahead
- Code for real scientific research problemsSciCode+13points ahead
Where Llama 3.3 70B Instruct pulls ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+6.1points ahead
- Common-sense trick questionsSimpleBench+1points ahead
Every test, side by side
All 34 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V3
- SciCodeDeepSeek-V3 by 133926+13
- Terminal-Bench HardDeepSeek-V3 by 12.215.23+12.2
- Terminal-Bench 2.1DeepSeek-V3 by 1216.94.9+12
ReasoningDeepSeek-V3
- GPQA DiamondDeepSeek-V3 by 15.765.549.8+15.7
- Humanity's Last ExamDeepSeek-V3 by 1.14.73.6+1.1
- SimpleBenchLlama 3.3 70B Instruct by 118.919.9+1
- CritPttie00tie
FactsDeepSeek-V3
- AA-Omniscience · AccuracyDeepSeek-V3 by 6.525.418.9+6.5
- AA-Omniscience · Non-hallucinationDeepSeek-V3 by 4.314.19.8+4.3
- Vectara HHEM hallucination ratelower is betterLlama 3.3 70B Instruct by 26.14.1+2
Following instructionsLlama 3.3 70B Instruct
- IFBenchLlama 3.3 70B Instruct by 6.14147.1+6.1
Other results20 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.
- HumanEvalLlama 3.3 70B Instruct by 23.265.288.4+23.2
- τ²-Bench Telecom (AA run)DeepSeek-V3 by 20.547.126.6+20.5
- vectara_avg_summary_lengthDeepSeek-V3 by 17.181.764.6+17.1
- LiveCodeBenchDeepSeek-V3 by 15.949.233.3+15.9
- AA-OmniscienceDeepSeek-V3 by 13.5-40.7-54.2+13.5
- MATHDeepSeek-V3 by 13.290.277+13.2
- MGSMLlama 3.3 70B Instruct by 11.379.891.1+11.3
- Artificial Analysis Coding IndexDeepSeek-V3 by 11.12311.9+11.1
- MMLU-ProDeepSeek-V3 by 775.968.9+7
- IFEvalLlama 3.3 70B Instruct by 686.192.1+6
- SimpleQADeepSeek-V3 by 424.920.9+4
- naturalquestions_closedbookDeepSeek-V3 by 3.646.743.1+3.6
- OpenBookQADeepSeek-V3 by 2.695.492.8+2.6
- MMLUDeepSeek-V3 by 2.588.586+2.5
- AA IntelligenceDeepSeek-V3 by 29.77.7+2
- vectara_answer_rateLlama 3.3 70B Instruct by 297.599.5+2
- vectara_factual_consistencyLlama 3.3 70B Instruct by 293.995.9+2
- DROPDeepSeek-V3 by 1.491.690.2+1.4
- NarrativeQAtie79.679.1tie
- GSM8Ktie9494.2tie
Questions people ask
Which is better, DeepSeek-V3 or Llama 3.3 70B Instruct?
DeepSeek-V3 wins five of the six areas where both have results: coding, agents, reasoning, facts and long documents. Llama 3.3 70B Instruct wins following instructions.
Which is better for coding?
DeepSeek-V3. It wins 3 of the 3 coding tests both models report; Llama 3.3 70B Instruct wins none.
Which is cheaper?
DeepSeek-V3 costs $0.24 per million input tokens and $0.90 per million output tokens; Llama 3.3 70B Instruct costs $0.71 and $0.72. That makes DeepSeek-V3 about 20% cheaper for the same work.
How do you compare the two?
We use the 34 benchmark tests both models have published scores on. The verdict counts the 14 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 20 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.