DeepSeek-V4-Pro 0813 vs GPT-5.5
Wins 1 of 7 areas
Agents
Wins 6 of 7 areas
Coding · Reasoning · Facts · Math · Long documents · Following instructions
GPT-5.5 is the stronger all-rounder.DeepSeek-V4-Pro 0813 is cheaper and better at agents.
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.
- ReasoningHard problems that need careful thinking04GPT-5.54 of 5 tests · 1 tie
- FactsGetting facts right instead of making them up03GPT-5.53 of 3 tests
- CodingWriting and fixing software13GPT-5.53 of 4 tests
- MathCompetition and research-level math02GPT-5.52 of 3 tests · 1 tie
- Long documentsFinding answers in very long texts01GPT-5.51 of 1 test
- Following instructionsDoing exactly what it is asked01GPT-5.51 of 1 test
- AgentsCarrying out multi-step tasks on its own10DeepSeek-V4-Pro 08131 of 3 tests · 2 ties
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-Pro 0813 costs 85% 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 GPT-5.5 pulls ahead
- Extremely hard research-level math problemsFrontierMath Tier 4+45.7points ahead
- Abstract visual puzzles that people can solveARC-AGI-2+23.7points ahead
- Unpublished advanced math problemsFrontierMath Tiers 1-3 (v2)+20.7points ahead
Where DeepSeek-V4-Pro 0813 pulls ahead
- Real work tasks from 44 professionsGDPVal+5points ahead
- Fixes real GitHub issues, working as an agentLiveBench · Agentic Coding+1points 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.
CodingGPT-5.5
- Terminal-Bench 2.1GPT-5.5 by 5.678.784.3+5.6
- LiveBench · CodingGPT-5.5 by 577.282.2+5
- SciCodeGPT-5.5 by 4.85155.8+4.8
- LiveBench · Agentic CodingDeepSeek-V4-Pro 0813 by 15554+1
AgentsDeepSeek-V4-Pro 0813
- GDPValDeepSeek-V4-Pro 0813 by 547.542.5+5
- τ-Bench V3 · Bankingtie39.639tie
- Terminal-Bench 4.0tie14.114.6tie
ReasoningGPT-5.5
- ARC-AGI-2GPT-5.5 by 23.761.385+23.7
- CritPtGPT-5.5 by 9.11827.1+9.1
- Humanity's Last ExamGPT-5.5 by 4.84145.8+4.8
- LiveBench · ReasoningGPT-5.5 by 3.985.889.7+3.9
- GPQA Diamondtie92.893.5tie
FactsGPT-5.5
- SimpleQA VerifiedGPT-5.5 by 10.152.963+10.1
- AA-Omniscience · AccuracyGPT-5.5 by 8.949.158+8.9
- AA-Omniscience · Non-hallucinationGPT-5.5 by 5.85.211+5.8
MathGPT-5.5
- FrontierMath Tier 4GPT-5.5 by 45.726.872.5+45.7
- FrontierMath Tiers 1-3 (v2)GPT-5.5 by 20.764.685.3+20.7
- LiveBench · Mathematicstie95.195.9tie
Following instructionsGPT-5.5
- LiveBench · Instruction FollowingGPT-5.5 by 367.770.7+3
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.
- AA-OmniscienceGPT-5.5 by 19.70.820.5+19.7
- ToolathlonDeepSeek-V4-Pro 0813 by 18.574.155.6+18.5
- NL2RepoDeepSeek-V4-Pro 0813 by 10.861.550.7+10.8
- GDPval-AA v2DeepSeek-V4-Pro 0813 by 81 rating points15901509+81 rating
- HLE (with tools)DeepSeek-V4-Pro 0813 by 7.86052.2+7.8
- DeepSWEGPT-5.5 by 7.362.770+7.3
- Artificial Analysis Coding IndexGPT-5.5 by 6.168.874.9+6.1
- livebench_languageGPT-5.5 by 5.382.187.4+5.3
- AA Agentic IndexDeepSeek-V4-Pro 0813 by 542.337.3+5
- ARC-AGI-1GPT-5.5 by 59095+5
- AA IntelligenceGPT-5.5 by 2.43638.4+2.4
- livebench_data_analysisGPT-5.5 by 2.479.281.6+2.4
- CyberGymDeepSeek-V4-Pro 0813 by 1.583.381.8+1.5
- Agents' Last Examtie25.726.6tie
- Toolathlon Verifiedtie74.173.5tie
Questions people ask
Which is better, DeepSeek-V4-Pro 0813 or GPT-5.5?
GPT-5.5 wins six of the seven areas where both have results: coding, reasoning, facts, math, long documents and following instructions. DeepSeek-V4-Pro 0813 wins agents, and costs 85% less.
Which is better for coding?
GPT-5.5. It wins 3 of the 4 coding tests both models report; DeepSeek-V4-Pro 0813 wins 1.
Which is cheaper?
DeepSeek-V4-Pro 0813 costs $1.32 per million input tokens and $3.96 per million output tokens; GPT-5.5 costs $5.00 and $30.00. That makes DeepSeek-V4-Pro 0813 about 85% 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 20 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.