Composer 2vsComposer 2.5

Composer 2
Composer 2.5
Benchmarks
SWE-Bench Multilingual
73.7%79.8%
Next.js Evals
58%81%
Terminal-Bench 2.0
61.7%69.3%
Benchmarks
SWE-Bench Pro
54%
DeepSWE 1.0
18%
Terminal-Bench 2.1
73%
threejseval
1137
Overview
CompanySpaceXAISpaceXAI
Release dateMar 19 2026May 18 2026
AccessProprietaryProprietary

Other comparisons

Composer 2vsClaude Fable 5.1Composer 2.5vsClaude Fable 5.1Composer 2vsGPT-6 AstraComposer 2.5vsGPT-6 AstraComposer 2vsGemini 3.8 FlashComposer 2.5vsGemini 3.8 FlashComposer 2vsMuse Spark 1.3Composer 2.5vsMuse Spark 1.3Composer 2vsDeepSeek-V4.1-FlashComposer 2.5vsDeepSeek-V4.1-FlashComposer 2vsMistral Medium 3.5Composer 2.5vsMistral Medium 3.5

Frequently asked questions

Composer 2.5 leads Composer 2 on 3 of the 3 benchmarks they both report (SWE-Bench Multilingual, Next.js Evals, Terminal-Bench 2.0). Composer 2 shipped 60 days before Composer 2.5, so benchmark comparisons should account for the intervening progress.

Published specifications for these two models are limited — see each model page for the latest details.

On SWE-Bench Multilingual, Composer 2.5 leads at 79.8% vs Composer 2 at 73.7%. On Next.js Evals, Composer 2.5 leads at 81% vs Composer 2 at 58%. On Terminal-Bench 2.0, Composer 2.5 leads at 69.3% vs Composer 2 at 61.7%.