o4-mini

LLM Provider / Model · Cheap fast reasoning, agentic coding loops, high-volume tasks

At a glance

Input price	$1.10
Output price	$4.40
Cache price	$0.275
Price tier	Budget
Context window	200K
Max output	100K
Context tier	128K-500K
Speed tier	Fast
Latency	fast
Knowledge cutoff	Jun 2024
Modality	Multimodal (vision)
Model ID	o4-mini
Provider	OpenAI
HumanEval	93%
MMLU	87%
SWE-Bench	68%
Benchmark	strong coding
Released	Apr 2025
Hosting	Closed/API
Capabilities	Vision, Tool use, Reasoning, Streaming, Structured output

Vision, Tool use, Reasoning, Streaming, Structured output

Cheap fast reasoning, agentic coding loops, high-volume tasks

Trae's built-in model menu may not include o4-mini — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
Google Antigravity is built around Gemini 3 Pro; Claude and recent GPT-5 models also work, but o4-mini is not in the supported model list.
Kiro's spec mode is tuned for Claude models — o4-mini may work but spec-driven flows are best with Claude.
Gemini CLI is locked to Google Gemini models — your LLM is o4-mini.
Claude Code only works with Claude models — your LLM is o4-mini.
Devin uses Cognition's proprietary model internally — your selected o4-mini will be ignored.
OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected o4-mini will be ignored.
Jules runs Gemini models internally — your selected o4-mini will be ignored.
The Claude Code SDK is tuned for Claude models — o4-mini may work via the model adapter but loses Claude-specific features (thinking, prompt caching, sub-agents).
Lovable uses a fixed internal model — your selected o4-mini will be ignored.
Replit Agent 3 uses its own model selection internally — your selected o4-mini will be ignored.

Build a full stack around o4-mini — Flowpicker shows compatibility warnings before you commit.