GPT-5.1-Codex-Max

Name: GPT-5.1-Codex-Max
Author: OpenAI

OpenAI · 2025-11-19

GPT-5.1-Codex-Max is OpenAI's frontier agentic coding model, optimized for long-running tasks. It was OpenAI's first model natively trained to operate across multiple context windows through compaction, and it supports variable reasoning effort including Extra High (xhigh).

Benchmark scores

Benchmark	Score
ALE-Bench	1538
App-Bench	38.4
BuseyBench SVG	4.9
Capability	151.8
IOI	21.4
IOI - IOI 2024	14.5
IOI - IOI 2025	28.3
LiveCodeBench (Vals AI public dataset)	83.6
LiveCodeBench (Vals Public) - Easy	97.5
LiveCodeBench (Vals Public) - Hard	65.4
LiveCodeBench (Vals Public) - Medium	87.7
Market-Bench	4243
Market-Bench - Dynamic Delta Hedging (Best of 5)	1371
Market-Bench - Dynamic Delta Hedging (Mean of 5)	10496
Market-Bench - Pairs Mean-Reversion (Best of 5)	89.4
Market-Bench - Pairs Mean-Reversion (Mean of 5)	137
Market-Bench - Single-Stock Scheduled Execution (Best of 5)	0.0
Market-Bench - Single-Stock Scheduled Execution (Mean of 5)	845
METR Time Horizons (1.1, 50% Success)	224
ProofBench	9.0
SnakeBench Average Apples	12.7
SnakeBench Best Apples	30.0
SnakeBench Rating	38.2
SnakeBench Total Apples	419
SnakeBench Win Rate	97.0
Terminal-Bench 2.0	60.4
Vibe Code Bench v1.1	22.2