DeepSeek-V4-Flash vs GPT-5
Wins 6 of 7 areas
Coding · Agents · Reasoning · Math · Long documents · Following instructions
Wins 1 of 7 areas
Facts
DeepSeek-V4-Flash is the stronger all-rounder.GPT-5 is better at facts.
Scores updated · 35 tests both models report · How we compare
Where each one wins
Tests won in each of the seven 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 software80DeepSeek-V4-Flash8 of 8 tests
- ReasoningHard problems that need careful thinking41DeepSeek-V4-Flash4 of 5 tests
- AgentsCarrying out multi-step tasks on its own30DeepSeek-V4-Flash3 of 3 tests
- MathCompetition and research-level math30DeepSeek-V4-Flash3 of 3 tests
- Long documentsFinding answers in very long texts10DeepSeek-V4-Flash1 of 1 test
- Following instructionsDoing exactly what it is asked10DeepSeek-V4-Flash1 of 1 test
- FactsGetting facts right instead of making them up02GPT-52 of 3 tests · 1 tie
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-V4-Flash costs 84% less for the same work.
The biggest differences
The three tests each model wins by the widest margin. Scores are out of 100.
Where DeepSeek-V4-Flash pulls ahead
- Abstract visual puzzles that people can solveARC-AGI-2+51.5points ahead
- Command-line tasks in a real terminalTerminal-Bench 2.1+43.5points ahead
- Real work tasks from 44 professionsGDPVal+25.9points ahead
Where GPT-5 pulls ahead
- Short factual questions, answered correctlySimpleQA Verified+16points ahead
- Common-sense trick questionsSimpleBench+10.4points ahead
- Avoids making up answers it doesn't knowAA-Omniscience · Non-hallucination+9.5points ahead
Every test, side by side
All 35 tests both models report. The winning score is in its model's colour; marks a score checked independently.
CodingDeepSeek-V4-Flash
- Terminal-Bench 2.1DeepSeek-V4-Flash by 43.578.735.2+43.5
- SWE-bench MultilingualDeepSeek-V4-Flash by 1873.355.3+18
- SWE-bench ProDeepSeek-V4-Flash by 10.852.641.8+10.8
- SciCodeDeepSeek-V4-Flash by 7.450.342.9+7.4
- SWE-bench VerifiedDeepSeek-V4-Flash by 4.17974.9+4.1
- LiveCodeBench v6DeepSeek-V4-Flash by 3.690.687+3.6
- Terminal-Bench HardDeepSeek-V4-Flash by 335.632.6+3
- LMArena · WebDevDeepSeek-V4-Flash by 13 rating points14301417+13 rating
AgentsDeepSeek-V4-Flash
- GDPValDeepSeek-V4-Flash by 25.946.921+25.9
- BrowseCompDeepSeek-V4-Flash by 18.373.254.9+18.3
- τ-Bench V3 · BankingDeepSeek-V4-Flash by 17.339.422.1+17.3
ReasoningDeepSeek-V4-Flash
- ARC-AGI-2DeepSeek-V4-Flash by 51.561.49.9+51.5
- CritPtDeepSeek-V4-Flash by 10.916.65.7+10.9
- SimpleBenchGPT-5 by 10.446.356.7+10.4
- Humanity's Last ExamDeepSeek-V4-Flash by 10.138.628.5+10.1
- GPQA DiamondDeepSeek-V4-Flash by 5.490.885.4+5.4
FactsGPT-5
- SimpleQA VerifiedGPT-5 by 1634.150.1+16
- AA-Omniscience · Non-hallucinationGPT-5 by 9.58.317.8+9.5
- AA-Omniscience · Accuracytie40.440.3tie
MathDeepSeek-V4-Flash
- IMOAnswerBenchDeepSeek-V4-Flash by 12.488.476+12.4
- FrontierMath Tier 4DeepSeek-V4-Flash by 2.424.421.9+2.4
- FrontierMath Tiers 1-3 (v2)DeepSeek-V4-Flash by 2.157.555.4+2.1
Long documentsDeepSeek-V4-Flash
- AA-LCRDeepSeek-V4-Flash by 1.579.778.2+1.5
Following instructionsDeepSeek-V4-Flash
- IFBenchDeepSeek-V4-Flash by 6.179.273.1+6.1
Other results11 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.
- Artificial Analysis Coding IndexDeepSeek-V4-Flash by 31.369.137.8+31.3
- ARC-AGI-1DeepSeek-V4-Flash by 23.38965.7+23.3
- Terminal-Bench 2.0DeepSeek-V4-Flash by 21.756.935.2+21.7
- AA Agentic IndexDeepSeek-V4-Flash by 15.241.726.5+15.2
- AA IntelligenceDeepSeek-V4-Flash by 11.334.323+11.3
- τ²-Bench Telecom (AA run)DeepSeek-V4-Flash by 10.29584.8+10.2
- HLE (with tools)DeepSeek-V4-Flash by 9.945.135.2+9.9
- LiveCodeBenchDeepSeek-V4-Flash by 8.191.683.5+8.1
- AA-OmniscienceGPT-5 by 5.6-14.3-8.7+5.6
- GPQA (unspecified)DeepSeek-V4-Flash by 2.888.585.7+2.8
- MMLU-Protie86.287.1tie
Questions people ask
Which is better, DeepSeek-V4-Flash or GPT-5?
DeepSeek-V4-Flash wins six of the seven areas where both have results: coding, agents, reasoning, math, long documents and following instructions. GPT-5 wins facts.
Which is better for coding?
DeepSeek-V4-Flash. It wins 8 of the 8 coding tests both models report; GPT-5 wins none.
Which is cheaper?
DeepSeek-V4-Flash costs $0.44 per million input tokens and $1.32 per million output tokens; GPT-5 costs $1.25 and $10.00. That makes DeepSeek-V4-Flash about 84% cheaper for the same work.
How do you compare the two?
We use the 35 benchmark tests both models have published scores on. The verdict counts the 24 tests in the eight capability areas, and a gap under one point (ten on rating-style scales) counts as a tie. The other 11 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.