GPT-5.1 Codex

LLM Provider / Model · Codex CLI and long-horizon coding agents; engineered for terminal-driven workflows

At a glance

Input price	$1.25
Output price	$10
Cache price	$0.125
Price tier	Mid
Context window	400K
Max output	128K
Context tier	128K-500K
Speed tier	Medium
Latency	medium
Knowledge cutoff	Oct 2025
Modality	Text + Vision
Model ID	gpt-5.1-codex
Provider	OpenAI
HumanEval	95%
MMLU	88%
SWE-Bench	75%
Benchmark	SWE-bench Verified ~75%
Released	Nov 2025
Hosting	Closed/API
Capabilities	Tool use, Coding-specialized fine-tune, Long-running tasks, Streaming, Function calling, Prompt caching

Tool use, Coding-specialized fine-tune, Long-running tasks, Streaming, Function calling, Prompt caching

Codex CLI and long-horizon coding agents; engineered for terminal-driven workflows

Kiro's spec mode is tuned for Claude models — GPT-5.1 Codex may work but spec-driven flows are best with Claude.
Gemini CLI is locked to Google Gemini models — your LLM is GPT-5.1 Codex.
Claude Code only works with Claude models — your LLM is GPT-5.1 Codex.
Devin uses Cognition's proprietary model internally — your selected GPT-5.1 Codex will be ignored.
OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected GPT-5.1 Codex will be ignored.
Jules runs Gemini models internally — your selected GPT-5.1 Codex will be ignored.
The Claude Code SDK is tuned for Claude models — GPT-5.1 Codex may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
Lovable uses a fixed internal model — your selected GPT-5.1 Codex will be ignored.
Replit Agent 3 uses its own model selection internally — your selected GPT-5.1 Codex will be ignored.

Build a full stack around GPT-5.1 Codex — Flowpicker shows compatibility warnings before you commit.