GPT-5.6 Luna

LLM Provider / Model · High-volume, cost-sensitive workloads that still need a large context window

At a glance

Input price	$0.20
Output price	$1.20
Cache price	$0.02
Price tier	Budget
Context window	1M+
Max output	128K
Context tier	500K+
Speed tier	Fast
Latency	fast
Knowledge cutoff	Feb 2026
Modality	Multimodal (vision)
Model ID	gpt-5.6-luna
Provider	OpenAI
Benchmark	Cheapest 1M-context model in the GPT-5.6 family
Released	Jul 2026
Hosting	Closed/API
Capabilities	Vision, Tool use, Reasoning effort control, Streaming, Structured output, Prompt caching

Vision, Tool use, Reasoning effort control, Streaming, Structured output, Prompt caching

High-volume, cost-sensitive workloads that still need a large context window

Trae's built-in model menu may not include GPT-5.6 Luna — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
Kiro's spec mode is tuned for Claude models — GPT-5.6 Luna may work but spec-driven flows are best with Claude.
Gemini CLI is locked to Google Gemini models — your LLM is GPT-5.6 Luna.
Claude Code only works with Claude models — your LLM is GPT-5.6 Luna.
Devin uses Cognition's proprietary model internally — your selected GPT-5.6 Luna will be ignored.
OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected GPT-5.6 Luna will be ignored.
Jules runs Gemini models internally — your selected GPT-5.6 Luna will be ignored.
The Claude Code SDK is tuned for Claude models — GPT-5.6 Luna 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-5.6 Luna will be ignored.
Replit Agent 3 uses its own model selection internally — your selected GPT-5.6 Luna will be ignored.

Build a full stack around GPT-5.6 Luna — Flowpicker shows compatibility warnings before you commit.