GPT-5.4 nano
LLM Provider / Model · Simple high-volume classification and extraction
At a glance
| Input price | $0.20 |
| Output price | $1.25 |
| Cache price | $0.02 |
| Price tier | Budget |
| Context window | 400K |
| Max output | 128K |
| Context tier | 128K-500K |
| Speed tier | Fast |
| Latency | fast |
| Knowledge cutoff | Aug 2025 |
| Modality | Multimodal (vision) |
| Model ID | gpt-5.4-nano |
| Provider | OpenAI |
| Benchmark | Cheapest OpenAI reasoning-family model |
| Released | Mar 2026 |
| Hosting | Closed/API |
| Capabilities | Streaming, Structured output, Tool use, Prompt caching |
What GPT-5.4 nano does
Streaming, Structured output, Tool use, Prompt caching
Best for
Simple high-volume classification and extraction
Works well with
Integration
Agent / Orchestration
Conflicts & caveats
- Trae's built-in model menu may not include GPT-5.4 nano — Trae primarily ships with Claude/GPT/Gemini/DeepSeek.
- Kiro's spec mode is tuned for Claude models — GPT-5.4 nano may work but spec-driven flows are best with Claude.
- Gemini CLI is locked to Google Gemini models — your LLM is GPT-5.4 nano.
- Claude Code only works with Claude models — your LLM is GPT-5.4 nano.
- Devin uses Cognition's proprietary model internally — your selected GPT-5.4 nano will be ignored.
- OpenAI Codex (Cloud) runs GPT-5.1 Codex internally — your selected GPT-5.4 nano will be ignored.
- Jules runs Gemini models internally — your selected GPT-5.4 nano will be ignored.
- The Claude Code SDK is tuned for Claude models — GPT-5.4 nano 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-5.4 nano will be ignored.
- Replit Agent 3 uses its own model selection internally — your selected GPT-5.4 nano will be ignored.
Build a full stack around GPT-5.4 nano — Flowpicker shows compatibility warnings before you commit.
Open the stack planner →