The Model Index
Up to four models

GPT-6 Astra vs DeepSeek-V4-Pro

On the index (GPT-4 = 100), GPT-6 Astra scores 193 and DeepSeek-V4-Pro 163. Side by side on the same third-party tests, with price and context. The link in your address bar shares this exact comparison.

Next to GPT-6 Astra
Presets
GPT-6 Astra
193178–209
WeightsClosed
ReleasedSep 2026
Price in / out$10 / $50
Context1.05M
Core tests taken7 of 9
DeepSeek-V4-Pro
163147–178
WeightsOpen
ReleasedAug 2026
Price in / out$0.11 / $4.44
Context1.05M
Core tests taken6 of 9

GPT-6 Astra leads DeepSeek-V4-Pro by 31 points, ahead on 6 of 6 shared core tests, at 17× the price.

Test by testthird-party results

GPT-6 AstraDeepSeek-V4-Pro
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)LiveBench Data AnalysisLiveBench Data AnalysisLiveBench IFLiveBench IFLiveBench LanguageLiveBench LanguageLiveBench ReasoningLiveBench ReasoningSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)LiveBench MathematicsLiveBench MathematicsProofBench v1.1ProofBench v1.1Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGTerminal-Bench 4.0Terminal-Bench 4.0LiveBench Agentic CodingLiveBench Agentic CodingLiveBench CodingLiveBench CodingCode MigrationCode MigrationCyberBench v1.1CyberBench v1.1IOIIOIProgramBenchProgramBenchVibe Code Bench 1-100Vibe Code Bench 1-100Vibe Code Bench v1.1Vibe Code Bench v1.1MirrorCodeMirrorCodeWeirdMLWeirdMLAGENTSAPEX-AgentsAPEX-AgentsSRE BenchSRE BenchTerminal-Bench-ScienceTerminal-Bench-ScienceDTBenchDTBenchLMCALMCAMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKFinance Agent (v2)Finance Agent (v2)EMBEMBHLAB-Vals-1.1HLAB-Vals-1.1Legal Research BenchLegal Research BenchMedCodeMedCodeMedScribeMedScribeTax Agent BenchTax Agent BenchNOVEL PROBLEMSARC-AGI-2ARC-AGI-2ARC-AGI-1ARC-AGI-1
Every result as a table
TestGPT-6 AstraDeepSeek-V4-Pro
Reasoning & knowledge
GPQA Diamond (Epoch AI)95.8%91.7%
LiveBench Data Analysis83.0%79.2%
LiveBench IF75.6%67.7%
LiveBench Language89.4%82.1%
LiveBench Reasoning92.7%85.8%
SimpleQA Verified75.6%52.9%
Maths
FrontierMath (tiers 1-3)93.7%64.6%
AIME-style maths (OTIS mock)100.0%98.6%
LiveBench Mathematics96.8%95.1%
ProofBench v1.199.0%50.0%
Chess puzzles74.0%46.0%
FrontierMath (tier 4)97.6%26.8%
Coding
Terminal-Bench 4.059.6%14.1%
LiveBench Agentic Coding57.3%54.9%
LiveBench Coding80.4%77.2%
Code Migration67.7%41.5%
CyberBench v1.141.1%72.9%
IOI100.0%51.6%
ProgramBench5.5%0.0%
Vibe Code Bench 1-10027.6%17.5%
Vibe Code Bench v1.189.6%82.3%
MirrorCode60.0%33.3%
WeirdML93.3%66.2%
Agents
APEX-Agents64.7%47.3%
SRE Bench56.9%1.9%
Terminal-Bench-Science62.9%4.3%
DTBench97.3%93.9%
LMCA64.4%45.5%
Mystery game puzzles84.0%43.0%
Professional work
Finance Agent (v2)53.5%50.4%
EMB71.7%52.8%
HLAB-Vals-1.17.1%10.8%
Legal Research Bench39.4%40.9%
MedCode48.5%42.5%
MedScribe87.9%80.2%
Tax Agent Bench20.7%26.8%
Novel problems
ARC-AGI-295.0%61.3%
ARC-AGI-198.5%90.5%

Tests at least two of these models have taken. Bold rows are the core tests.

This page as a chartDownload PNG