gpt-5-nanoOpenAI
- endpoints
- 8
- regions
- 6
- api
- chat
- released
- August 2025
- eu routing
- available
capabilities 5/8
The same weights cost 2.2x more on microsoft azure ai than on openai inc.. That is what the endpoints table is for.
| # | flags | ||||||
|---|---|---|---|---|---|---|---|
| 1 |  OpenAI Inc. | Global | 400K | $0.02 | $0.20 | $0.0025 | no-train |
| 2 |  Microsoft Azure AI | Global | 200K | $0.05 | $0.40 | $0.0050 | zdr |
| 3 | OpenAI Inc. | Global | 400K | $0.05 | $0.40 | $0.0050 | no-train |
| 4 | Microsoft Azure AI | eastus2 | 200K | $0.05 | $0.40 | $0.0050 | zdr |
| 5 | Microsoft Azure AI | francecentral | 200K | $0.06 | $0.44 | $0.0055 | zdr |
| 6 | Microsoft Azure AI | germanywestcentral | 200K | $0.06 | $0.44 | $0.0055 | zdr |
| 7 | Microsoft Azure AI | swedencentral | 200K | $0.06 | $0.44 | $0.0055 | zdr |
| 8 | Microsoft Azure AI | westeurope | 200K | $0.06 | $0.44 | $0.0055 | zdr |
no measured traffic for this model yet. the endpoints table above carries provider pricing
| eval | Score as a share of 100 | /100 |
|---|---|---|
| Math Index | 83.7 | |
| AIME 2025 | 83.7 | |
| LiveCodeBench | 78.9 | |
| MMLU Pro | 78.0 | |
| GPQA Diamond | 67.6 | |
| τ²-Bench | 36.5 | |
| Intelligence Index | 13.0 | |
| Terminal-Bench Hard | 12.1 | |
| Humanity's Last Exam | 9.5 | |
| 9 evals | mean | 51.4 |
Scores from Artificial Analysis and public leaderboards, normalised to 100. Bar colour is the band, not the rank: green 80 and up, blue 55 and up, amber 30 and up. Benchmarks measure narrow skills, so test on your own workload before committing. Released 2025-08-07.
gpt-5-nano@eusame routing, restricted to eu-hosted providers. use this id when data residency is a requirement; requests never leave the eu
Base url is https://router.requesty.ai/v1 for every id here. One key reaches the whole catalog.
from openai import OpenAI client = OpenAI( api_key="YOUR_REQUESTY_API_KEY", base_url="https://router.requesty.ai/v1",) response = client.chat.completions.create( model="openai/gpt-5-nano:flex", messages=[ {"role": "user", "content": "Explain quantum computing in one paragraph."}, ],) print(response.choices[0].message.content)
Change the base url, use your Requesty key, set the model to any id in the call pane. Existing OpenAI SDK code needs no other edit, and the same key reaches every other model in the catalog, so switching later is a one-parameter change. Browse all models
GPT-5 nano is OpenAI's fastest, cheapest version of GPT-5. It's great for summarization and classification tasks.
questions 5
Which providers serve gpt-5-nano?
How much does gpt-5-nano cost?
Is gpt-5-nano open weights?
What is the context window of gpt-5-nano?
How do I use gpt-5-nano with the OpenAI SDK?
more from openai 6
| model | context | from /M | providers |
|---|---|---|---|
| gpt-6-astra | 1.1M | $5.00 | 2 |
| gpt-5.6-luna | 1.1M | $0.10 | 3 |
| gpt-5.6-terra | 1.1M | $1.00 | 3 |
| gpt-5.6-sol | 1.1M | $2.50 | 3 |
| gpt-5.5-pro | 1.1M | $30.00 | 1 |
| gpt-5.5 | 1.1M | $2.50 | 3 |
route gpt-5-nano through one endpoint
One key for 2 providers on this model and 600+ others. No markup on provider prices, automatic failover, caching built in. Weekly aggregates in the measured pane come from production traffic routed through Requesty, one line per provider on a shared axis. Methodology