DeepSeek-V3 vs Llama 3.2 Instruct 11B (Vision)
Wins 4 of 5 areas
Coding · Reasoning · Long documents · Following instructions
Wins 0 of 5 areas
—
DeepSeek-V3 is the stronger all-rounder.Llama 3.2 Instruct 11B (Vision) is cheaper.
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 software20DeepSeek-V32 of 2 tests
- ReasoningHard problems that need careful thinking10DeepSeek-V31 of 3 tests · 2 ties
- Long documentsFinding answers in very long texts10DeepSeek-V31 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-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.
Llama 3.2 Instruct 11B (Vision) costs 39% 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
- Graduate-level biology, physics and chemistry questionsGPQA Diamond+43.4points ahead
- Reasons across sets of long documentsAA-LCR+28points ahead
- Code for real scientific research problemsSciCode+27.8points ahead
Where Llama 3.2 Instruct 11B (Vision) pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+4points 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
- SciCodeDeepSeek-V3 by 27.83911.2+27.8
- Terminal-Bench HardDeepSeek-V3 by 14.415.20.8+14.4
ReasoningDeepSeek-V3
- GPQA DiamondDeepSeek-V3 by 43.465.522.1+43.4
- Humanity's Last Examtie4.75.5tie
- CritPttie00tie
FactsEven
- AA-Omniscience · AccuracyDeepSeek-V3 by 14.925.410.6+14.9
- AA-Omniscience · Non-hallucinationLlama 3.2 Instruct 11B (Vision) by 414.118.1+4
Following instructionsDeepSeek-V3
- IFBenchDeepSeek-V3 by 10.64130.4+10.6
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.
- MMLUDeepSeek-V3 by 37.488.551.1+37.4
- τ²-Bench Telecom (AA run)DeepSeek-V3 by 32.547.114.6+32.5
- naturalquestions_closedbookDeepSeek-V3 by 23.346.723.4+23.3
- OpenBookQADeepSeek-V3 by 2395.472.4+23
- AA-OmniscienceDeepSeek-V3 by 21.9-40.7-62.6+21.9
- Artificial Analysis Coding IndexDeepSeek-V3 by 18.8234.2+18.8
- MATHDeepSeek-V3 by 16.390.273.9+16.3
- GSM8KDeepSeek-V3 by 11.79482.3+11.7
- AA IntelligenceDeepSeek-V3 by 4.39.75.4+4.3
- NarrativeQADeepSeek-V3 by 479.675.6+4
Questions people ask
Which is better, DeepSeek-V3 or Llama 3.2 Instruct 11B (Vision)?
DeepSeek-V3 wins four of the five areas where both have results: coding, reasoning, long documents and following instructions. Llama 3.2 Instruct 11B (Vision) wins none, but costs 39% less. They are level on facts.
Which is better for coding?
DeepSeek-V3. It wins 2 of the 2 coding tests both models report; Llama 3.2 Instruct 11B (Vision) wins none.
Which is cheaper?
DeepSeek-V3 costs $0.24 per million input tokens and $0.90 per million output tokens; Llama 3.2 Instruct 11B (Vision) costs $0.34 and $0.34. That makes Llama 3.2 Instruct 11B (Vision) about 39% 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.