GPT-5.3 Codex Spark

Coding-optimized GPT model for repository edits, reviews, and agentic software work

BENCHMARKS

Not in the Epoch Capabilities Index. Epoch benchmarks what it can run, so this is a gap in coverage, not a poor result.

From the lab

Launch announcement · Feb 12, 2026Introducing GPT-5.3-Codex-Spark
RELEASE DATEFeb 5, 2026
WEIGHTSClosed weights
CONTEXT WINDOW128K
MAX OUTPUT32K
REASONINGYes
TOOL CALLINGYes
INPUTStext, image, pdf
OUTPUTStext

Specifications from Models.dev. Availability and limits may differ by provider. Benchmark scores come from Epoch AI under CC BY 4.0.