impulse

GPT-5 Nano

Frontier labRetiring 11 Dec 2026

GPT-5-Nano is the smallest and fastest variant in the GPT-5 system, optimized for developer tools, rapid interactions, and ultra-low latency environments.

API id gpt-5-nanoReleased 7 Aug 2025 (Aug 2025)
A newer model in this line is out: GPT-5.4 Nano, released 17 Mar 2026.
OpenAI retires it on 11 Dec 2026 (in 22 days); Azure OpenAI on 9 Feb 2027. 4 hosts keep serving it.
Input / 1M
$0.05
at OpenAI
Output / 1M
$0.40
at OpenAI
Context window
272K
tokens
Max output
128K
tokens per reply
Hosts
6
12 listings
Released
Aug 2025
7 Aug 2025

Pricing

List prices at OpenAI, in US dollars per million tokens.

TierInput / 1MOutput / 1M
Standard$0.05$0.40
Cached input
90% off input
$0.005
Batch
50% off
$0.025$0.20
Flex
50% off
$0.025$0.20
Web search
$10.00 per 1,000 searches

What it costs a month

At OpenAI list prices, 30 days.

  • Chat assistant5,000 requests a day, 2K in / 0.5K out, cached$40.95
  • RAG over documents2,000 requests a day, 8K in / 0.6K out, cached$27.60
  • Coding agent1,000 requests a day, 40K in / 2K out, cached$46.20
  • Overnight batch20,000 requests a day, 4K in / 0.4K out, batch$108
Use your own numbers

Where to run it

6 hosts serve GPT-5 Nano. The comparison is on blended price, 3 parts input to 1 part output.

HostIn / out per 1M
OpenAILab API
gpt-5-nano
$0.05 / $0.40
Retires 11 Dec 2026
databricks-gpt-5-nano
$0.05 / $0.40
gpt-5-nano
$0.05 / $0.40
Retires 9 Feb 2027
openai.gpt-5-nano
$0.05 / $0.40
ReplicateInference
gpt-5-nano
$0.05 / $0.40
PerplexityGateway
gpt-5-nano
$0.05 / $0.40
Show 6 more listings: regions, snapshots and variants
OpenAI
2025-08-07
gpt-5-nano-2025-08-07
$0.05 / $0.40
Retires 11 Dec 2026
Azure OpenAI
2025-08-07
gpt-5-nano-2025-08-07
$0.05 / $0.40
Retires 9 Feb 2027
eu
gpt-5-nano
$0.055 / $0.44
+10% vs headline
Retires 9 Feb 2027
us
gpt-5-nano
$0.055 / $0.44
+10% vs headline
Retires 9 Feb 2027
eu · 2025-08-07
gpt-5-nano-2025-08-07
$0.055 / $0.44
+10% vs headline
Retires 9 Feb 2027
us · 2025-08-07
gpt-5-nano-2025-08-07
$0.055 / $0.44
+10% vs headline
Retires 9 Feb 2027

Capabilities

Core

  • Streaming
  • Function calling
  • JSON mode
  • Structured outputs

Multimodal

  • Vision
  • Image input
  • Audio
  • Image generation
  • Video

Advanced

  • Reasoning
  • Tool calling
  • MCP
  • Prompt caching
  • Batch API
  • Embeddings
Takes in
ImagesTextPDFs
Produces
Text
Effort settings the source lists
minimal
Features
Parallel tool callsPDF inputPrompt cachingBuilt-in web search

Cheaper alternatives

Current models that cost less and keep most of what this one can do: tools, vision, reasoning, structured output, caching and enough context.

Same model, cheaper host

No host sells it for less than OpenAI.

Smaller OpenAI models

No cheaper OpenAI model keeps its key capabilities.

Other frontier labs

No cheaper match from another frontier lab.

Open-weight picks

  • MiniMax M3.1 Flash
    MiniMax · on Refuseless
    $0.05 / $0.25
    Has every key capability
    27% less
  • Qwen3.5 9B
    Alibaba Qwen · on EmpirioLabs AI
    $0.09 / $0.13
    Has every key capability
    27% less
  • GLM 5.3 Flash
    Zhipu AI · on CLF AI Gateway
    $0.06 / $0.20
    Has every key capability
    31% less
  • Gemma 4 26B A4B
    Google · on EmpirioLabs AI
    $0.05 / $0.29
    Has every key capability
    20% less

Model line

Newer and older versions, at each lab's own price.

  1. Mar 2026GPT-5.4 NanoLatest$0.20 / $1.25
  2. Aug 2025GPT-5 NanoThis page$0.05 / $0.40
  3. Apr 2025GPT-4.1 Nano$0.10 / $0.40

Call it

Using OpenAI's API.

Python
from openai import OpenAI

client = OpenAI()  # reads OPENAI_API_KEY

response = client.responses.create(model="gpt-5-nano", input="Hello")
print(response.output_text)

Quick answers

How much does GPT-5 Nano cost?

On OpenAI, GPT-5 Nano costs $0.05 per million input tokens and $0.40 per million output tokens. Cached input costs $0.005 per million tokens.

Which APIs offer GPT-5 Nano?

GPT-5 Nano runs on 6 APIs: OpenAI, Databricks, Azure OpenAI, Oracle OCI, Replicate and Perplexity.

What is GPT-5 Nano's context window?

272K tokens, with up to 128K tokens in one reply.

Is GPT-5 Nano being retired?

OpenAI retires it on 11 Dec 2026; Azure OpenAI on 9 Feb 2027. 4 hosts keep serving it.

Is there a newer version of GPT-5 Nano?

Yes. GPT-5.4 Nano is the newest model in this line, released on 17 Mar 2026.

What are cheaper alternatives to GPT-5 Nano?

Models that cost less and keep most of what it can do: MiniMax M3.1 Flash on Refuseless (27% less). Savings are on a mix of 3 input tokens to 1 output token.

When was GPT-5 Nano released?

OpenAI released it on 7 Aug 2025.

What is GPT-5 Nano's API model id?

Use gpt-5-nano with OpenAI. Other hosts use their own ids: databricks-gpt-5-nano on Databricks, openai.gpt-5-nano on Oracle OCI.

Sources

Raw catalog entry for gpt-5-nano on OpenAI
{
  "cache_read_input_token_cost": 5e-9,
  "cache_read_input_token_cost_batches": 2.5e-9,
  "cache_read_input_token_cost_flex": 2.5e-9,
  "deprecation_date": "2026-12-11",
  "deprecation_source": "https://developers.openai.com/api/docs/deprecations",
  "successor_model": "gpt-6-luna",
  "input_cost_per_token": 5e-8,
  "input_cost_per_token_batches": 2.5e-8,
  "input_cost_per_token_flex": 2.5e-8,
  "litellm_provider": "openai",
  "max_input_tokens": 272000,
  "max_output_tokens": 128000,
  "mode": "chat",
  "output_cost_per_token": 4e-7,
  "fedramp_uplift_multiplier": 1.1,
  "output_cost_per_token_batches": 2e-7,
  "output_cost_per_token_flex": 2e-7,
  "reasoning_contexts": [
    "auto",
    "current_turn"
  ],
  "reasoning_modes": [
    "standard"
  ],
  "search_context_cost_per_query": {
    "search_context_size_high": 0.01,
    "search_context_size_low": 0.01,
    "search_context_size_medium": 0.01
  },
  "source": "https://developers.openai.com/api/docs/pricing",
  "supported_endpoints": [
    "/v1/chat/completions",
    "/v1/batch",
    "/v1/responses"
  ],
  "supported_modalities": [
    "text",
    "image"
  ],
  "supported_output_modalities": [
    "text"
  ],
  "supports_function_calling": true,
  "supports_native_streaming": true,
  "supports_parallel_function_calling": true,
  "supports_pdf_input": true,
  "supports_prompt_caching": true,
  "supports_reasoning": true,
  "supports_response_schema": true,
  "supports_system_messages": true,
  "supports_tool_choice": true,
  "supports_vision": true,
  "supports_web_search": true,
  "supports_none_reasoning_effort": false,
  "supports_xhigh_reasoning_effort": false,
  "supports_minimal_reasoning_effort": true
}