DeepSeek-V3 wins more areas, narrowly.
Scores updated · 20 tests both models report · How we compare
Where each one wins
Tests won in each of the four 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.
Agents, reasoning 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 99% 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+30.6points ahead
- Follows unfamiliar, precisely checkable instructionsIFBench+7.8points ahead
- Answers hard knowledge questions correctlyAA-Omniscience · Accuracy+4.5points ahead
Where GPT-4 pulls ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+16.5points ahead
Every test, side by side
All 20 tests both models report. The winning score is in its model's colour; marks a score checked independently.
FactsEven
- AA-Omniscience · Non-hallucinationGPT-4 by 16.514.130.6+16.5
- AA-Omniscience · AccuracyDeepSeek-V3 by 4.525.421+4.5
Following instructionsDeepSeek-V3
- IFBenchDeepSeek-V3 by 7.84133.2+7.8
Other results15 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.
- MATHDeepSeek-V3 by 47.790.242.5+47.7
- AIR-Bench 2024GPT-4 by 23.440.864.2+23.4
- MMLUDeepSeek-V3 by 1588.573.5+15
- DROPDeepSeek-V3 by 10.791.680.9+10.7
- Artificial Analysis Coding IndexDeepSeek-V3 by 9.92313.1+9.9
- AA-OmniscienceGPT-4 by 6.9-40.7-33.8+6.9
- HellaSwagGPT-4 by 6.488.995.3+6.4
- MGSMDeepSeek-V3 by 5.379.874.5+5.3
- AA IntelligenceDeepSeek-V3 by 39.76.7+3
- NarrativeQADeepSeek-V3 by 2.879.676.8+2.8
- WinoGrandeGPT-4 by 2.684.987.5+2.6
- HumanEvalGPT-4 by 1.865.267+1.8
- naturalquestions_closedbookDeepSeek-V3 by 146.745.7+1
- GSM8Ktie9493.2tie
- OpenBookQAtie95.496tie
Questions people ask
Which is better, DeepSeek-V3 or GPT-4?
DeepSeek-V3 wins two of the four areas where both have results: reasoning and following instructions. GPT-4 wins none. They are level on agents and facts.
Which is cheaper?
DeepSeek-V3 costs $0.24 per million input tokens and $0.90 per million output tokens; GPT-4 costs $30.00 and $60.00. That makes DeepSeek-V3 about 99% cheaper for the same work.
How do you compare the two?
We use the 20 benchmark tests both models have published scores on. The verdict counts the 5 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 15 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.