The Model Index
Up to four models

GPT-6.2 vs GPT-6.2 Astra

On the index (GPT-4 = 100), GPT-6.2 scores 202 and GPT-6.2 Astra 202. 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-6.2
Presets
GPT-6.2
202184–220
WeightsClosed
ReleasedNov 2026
Price in / out$8.0 / $40
Context1.05M
Core tests taken3 of 9
GPT-6.2 Astra
202184–220
WeightsClosed
ReleasedNov 2026
Price in / out$8.0 / $40
Context1.05M
Core tests taken3 of 9

GPT-6.2 and GPT-6.2 Astra are too close to call (0 points apart, within the uncertainty), ahead on 0 of 3 shared core tests, at a similar price.

Test by testthird-party results

GPT-6.2GPT-6.2 Astra
0%25%50%75%100%REASONING & KNOWLEDGEGPQA Diamond (Epoch AI)GPQA Diamond (Epoch AI)EBR-benchEBR-benchSimpleQA VerifiedSimpleQA VerifiedMATHSFrontierMath (tiers 1-3)FrontierMath (tiers 1-3)AIME-style maths (OTIS mock)AIME-style maths (OTIS mock)Chess puzzlesChess puzzlesFrontierMath (tier 4)FrontierMath (tier 4)CODINGTerminal-Bench 4.0Terminal-Bench 4.0Vibe Code Bench v1.1Vibe Code Bench v1.1MirrorCodeMirrorCodeAGENTSBioMysteryBenchBioMysteryBenchTerminal-Bench-ScienceTerminal-Bench-ScienceFurniture assemblyFurniture assemblyMystery game puzzlesMystery game puzzlesPROFESSIONAL WORKHLAB-Vals-1.1HLAB-Vals-1.1Legal Research BenchLegal Research BenchTax Agent BenchTax Agent BenchNOVEL PROBLEMSMysteryMechanismMysteryMechanism
Every result as a table
TestGPT-6.2GPT-6.2 Astra
Reasoning & knowledge
GPQA Diamond (Epoch AI)96.8%96.8%
EBR-bench92.4%92.4%
SimpleQA Verified78.5%78.5%
Maths
FrontierMath (tiers 1-3)92.6%92.6%
AIME-style maths (OTIS mock)100.0%100.0%
Chess puzzles78.0%78.0%
FrontierMath (tier 4)100.0%100.0%
Coding
Terminal-Bench 4.063.1%63.1%
Vibe Code Bench v1.189.9%89.9%
MirrorCode67.8%67.8%
Agents
BioMysteryBench80.7%80.7%
Terminal-Bench-Science71.4%71.4%
Furniture assembly89.2%89.2%
Mystery game puzzles95.0%95.0%
Professional work
HLAB-Vals-1.116.3%16.3%
Legal Research Bench42.8%42.8%
Tax Agent Bench21.8%21.8%
Novel problems
MysteryMechanism53.1%53.1%

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

This page as a chartDownload PNG