GPT-5.4 Nano is a OpenAI model available through Merge Gateway. Use it with Gateway routing policies, spend controls, request logs, and a 272,000 token context window. It supports streaming, structured outputs, tool calling, vision through at least one Gateway vendor route.
GPT-5.4 Nano pricing
| Vendor | Input / 1M tokens | Output / 1M tokens | Zero data retention |
| --- | ---: | ---: | --- |
| OpenAI | $0.2000 | $1.25 | Yes |
Test GPT-5.4 Nano with Merge Gateway’s Simulator
GPT-5.4 Nano
Synced
Synced
Run simulation to see response
—
—
—
Ready to try it out?
Start routing requests to hundreds of large language models in your product within minutes.
Route requests to GPT-5.4 Nano with Merge Gateway
Merge Gateway is a unified LLM API that lets your product route requests to GPT-5.4 Nano and every other major model through a single endpoint. You get built-in fallback routing, per-request cost tracking, zero data retention support, and observability without changing your application architecture.
To get started in seconds, add our Gateway Implementation skill to your project, or pick your preferred SDK below. Check out our other quick start skills here.
Install the Merge Gateway SDK
Python
1$ pip install merge-gateway-sdkSend a request
Python
1from merge_gateway import MergeGateway
2
3client = MergeGateway(api_key="YOUR_API_KEY")
4
5response = client.responses.create(
6 model="openai/gpt-5.2",
7 input=[
8 {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
9 {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
10 ],
11)
12
13print(response.output[0].content[0].text)Try a diffrent model
Swap the model string to route to a different provider. No other code changes needed.
Anthropic
1response = client.responses.create(
2 model="anthropic/claude-sonnet-4-20250514",
3 input=[
4 {"type": "message", "role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5 {"type": "message", "role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6 ],
7)Point to Gateway
Python
1from openai import OpenAI
2
3client = OpenAI(
4 api_key="YOUR_API_KEY",
5 base_url="https://api-gateway.merge.dev/v1/openai",
6)Send a request
Use the standard chat.completions.create method. No provider prefix needed on the model name.
Python
1response = client.chat.completions.create(
2 model="gpt-5.2",
3 messages=[
4 {"role": "system", "content": "You are a helpful programming tutor. Explain the concepts clearly with practical examples."},
5 {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
6 ],
7)
8
9print(response.choices[0].message.content)Install packages
1npm install merge-gateway-ai-sdk-provider aiCreate the provider
TypeScript
1import { createMergeGateway } from "merge-gateway-ai-sdk-provider";
2
3const gateway = createMergeGateway({
4 apiKey: "YOUR_API_KEY",
5});Send a request
Use generateText to send a request. Model names use the provider/model format.
TypeScript
1import { generateText } from "ai";
2
3const { text } = await generateText({
4 model: gateway("openai/gpt-4o"),
5 prompt: "Explain the concept of recursion in programming with a simple set of examples.",
6});
7
8console.log(text);If you already have @ai-sdk/openai installed, point it at Gateway with a base URL change:
TypeScript
1import { createOpenAI } from "@ai-sdk/openai";
2
3const gateway = createOpenAI({
4 apiKey: "YOUR_API_KEY",
5 baseURL: "https://api-gateway.merge.dev/v1/ai-sdk",
6});
7
8// All generateText/streamText calls work unchangedInstall the Merge Gateway SDK
Anthropic SDK
1from anthropic import Anthropic
2
3client = Anthropic(
4 api_key="YOUR_API_KEY",
5 base_url="https://api-gateway.merge.dev/v1/anthropic",
6)
7
8message = client.messages.create(
9 model="claude-sonnet-4-20250514",
10 max_tokens=1024,
11 messages=[
12 {"role": "user", "content": "Explain the concept of recursion in programming with a simple set of examples."},
13 ],
14)
15
16print(message.content[0].text)
GPT-5.4 Nano FAQ
In case you have any other questions on GPT-5.4 Nano, we've answered a few more below. It's worth noting that the information below was written in June, 2026 and is subject to change.
Heading
What other models does OpenAI offer?
OpenAI's lineup spans cost-optimized, general-purpose, and frontier reasoning tiers. Here are some other models OpenAI supports:
- GPT-4.1 Nano: GPT-4.1 Nano is OpenAI's non-reasoning cost-tier model at $0.10 input and $0.40 output per million tokens with a 1M token context window. It delivers near-instant responses and is suited for high-volume tasks where reasoning is not required
- GPT-5.4 Mini: GPT-5.4 Mini is the next tier up from GPT-5.4 Nano in the same series, priced at $0.75 input and $4.50 output per million tokens. It scores 49 on the Intelligence Index versus GPT-5.4 Nano's 44 and is the right escalation step for tasks that need more capability
- GPT-5: GPT-5 is OpenAI's mid-range GPT-5 series model with a 400k token context window, offering higher intelligence than the Nano tier
- GPT-5.4: GPT-5.4 is the flagship in the GPT-5.4 series, scoring 57 on the Intelligence Index at $2.50 input and $15.00 output per million tokens. It delivers significantly higher capability than GPT-5.4 Nano for ceiling-level tasks
- GPT-5.5: GPT-5.5 is OpenAI's current highest-capability model at $5.00 input and $30.00 output per million tokens, ranked #2 globally on the Artificial Analysis Intelligence Index
How does GPT-5.4 Nano differ from OpenAI's other models?
GPT-5.4 Nano is OpenAI's entry-level offering in the GPT-5.4 series, combining reasoning capability with low per-token pricing and official API support.
- Pricing: GPT-5.4 Nano is priced at $0.20 per million input tokens and $1.25 per million output tokens. These figures are confirmed on OpenAI's official pricing page, making it a reliable option for production cost planning
- Intelligence ranking: GPT-5.4 Nano scores 44 on the Artificial Analysis Intelligence Index, placing it #10 out of 163 evaluated models in its class. This is notably higher than GPT-5 Nano at 27 and GPT-5 Mini at 41, despite GPT-5.4 Nano's lower price point
- Speed: Output speed is 149.7 tokens per second. Time to first token is 6.33 seconds, making it the fastest-responding model in the GPT-5.4 series and viable for near-real-time batch processing
- Context window: GPT-5.4 Nano supports 400,000 input tokens, consistent with the rest of the GPT-5.4 series
- Knowledge cutoff: GPT-5.4 Nano has a knowledge cutoff of August 2025, compared to May 2024 for GPT-5 Nano and GPT-5 Mini
- Capabilities: Accepts text and image inputs, produces text output. Reasoning model released March 17, 2026
GPT-5.4 Nano is the standout choice in the low-cost reasoning tier: it scores 44 on intelligence, costs $0.20 per million input tokens, has official OpenAI pricing, and delivers a 6.33-second TTFT that makes it usable for semi-real-time batch workflows.
What models should I consider using alongside GPT-5.4 Nano?
No single model is optimal for every task. Here are models worth pairing with GPT-5.4 Nano depending on what your product needs:
- GPT-5.4 Mini: When a task exceeds GPT-5.4 Nano's capability, escalate to GPT-5.4 Mini at $0.75 input per million tokens for a 5-point Intelligence Index improvement and still-reasonable pricing. Use it selectively to keep costs controlled
- GPT-4.1 Nano: For high-volume tasks that don't require reasoning traces and need the fastest possible response time or a larger context window, GPT-4.1 Nano delivers a 1M token context at $0.10 input with near-instant TTFT
- GPT-5.4: For ceiling-level tasks that exceed GPT-5.4 Nano's intelligence score, GPT-5.4 at $2.50 input per million tokens offers a 13-point Intelligence Index improvement. Reserve it for tasks where quality is the binding constraint
- Claude Haiku 4.5: For cross-provider diversity in the cost tier, Claude Haiku 4.5 at $1.00 input per million tokens provides a fast Anthropic alternative for high-volume workloads
- Gemini 2.5 Flash Lite: For latency-critical volume workloads, Gemini 2.5 Flash Lite delivers very high output speed at a comparable or lower cost tier, useful as a failover when OpenAI is rate-limited
What are the challenges of using GPT-5.4 Nano in my product?
Like any production LLM, GPT-5.4 Nano comes with tradeoffs worth planning for:
- TTFT still too high for synchronous UI: At 6.33 seconds time to first token, GPT-5.4 Nano is faster than other GPT-5.4 series models but still too slow for real-time chat interfaces or interactive streaming. It is suited to async and batch pipelines
- High output verbosity: GPT-5.4 Nano generated 210 million output tokens during evaluation, a very high figure relative to the dataset median. At $1.25 per million output tokens, verbose responses will meaningfully inflate production costs
- Smaller context window than non-reasoning alternatives: The 400,000 token context window is smaller than the 1M token context available on GPT-4.1 Nano and GPT-4.1 Mini. Applications with very long documents or extended conversations may need a different model
- Provider dependency: Running exclusively on OpenAI creates fragility from outages, rate limits, or future model deprecations. Anthropic, Google, and Mistral offer comparable cost-tier models as alternatives
- Cost at scale: Output verbosity combined with $1.25 per million output tokens means actual spend can be higher than the input-only pricing suggests, particularly for tasks that generate long responses
Why should I use Merge Gateway to route LLM requests with GPT-5.4 Nano and every other model?
Using GPT-5.4 Nano through Merge Gateway gives you access to the model itself and the infrastructure layer around it:
- One API, every provider: Access GPT-5.4 Nano and every other major LLM through a single endpoint and API key. Change providers by swapping the model string, with no application code changes required
- Intelligent routing and automatic failover: Merge routes around OpenAI outages automatically. Routing policies can automatically escalate to GPT-5.4 Mini for harder tasks or fall back to GPT-4.1 Nano when speed is the priority, without code changes
- Cost governance: Set hard or soft project budgets so GPT-5.4 Nano spend stays within plan. Given its output verbosity, budget tracking across projects and tags helps prevent unexpected cost spikes
- Build Your Own Router: Define the intelligence threshold at which GPT-5.4 Nano is sufficient versus when to escalate by scoring models against your own benchmark weights. The router picks the winner per request with a plain-language explanation of every decision
- Security and compliance controls: Apply DLP rules and prompt injection protection before every request reaches OpenAI. Enforce per-project model and region policies without adding that logic to your application
How can I start using Merge Gateway to route requests with GPT-5.4 Nano?
Getting GPT-5.4 Nano running through Merge Gateway takes a few minutes:
1. Create an account and get your API key from the dashboard.
2. Install the Merge Gateway SDK: run pip install merge-gateway-sdk (Python) or npm install merge-gateway-sdk (Node). Alternatively, if you're already using the OpenAI SDK, set base_url = "https://api-gateway.merge.dev/v1/openai" and your existing code works as-is.
3. Make your first request using the provider/model format. For GPT-5.4 Nano, the model string is openai/gpt-5.4-nano. Swap the model string to route to any other provider without changing anything else.
4. Configure a routing policy in the dashboard to set failover behavior, cost limits, and optimization strategy. A practical starting point: use GPT-5.4 Nano as the default and escalate to GPT-5.4 Mini for tasks where output quality falls short, keeping GPT-5.4 as a ceiling-level escalation option.
Full setup instructions and SDK references are in the Merge Gateway docs.
Try GPT-5.4 Nano through Merge Gateway
Route, observe, and control AI requests across providers from one API.
