For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million token context window, and scores 80.1% on MMLU, 50.3% on GPQA, and 9.8% on Aider polyglot coding – even higher than GPT‑4o mini. It’s ideal for tasks like classification or autocompletion.
Modalities
In / Out Price
$0.10 / $0.40per 1M
Context
1.0M
Released
Apr 14, 2025
Knowledge Cutoff
Jun 2024
Different companies host the same model. OpenRouter routes your request to one of them based on the routing mode you pick — Balanced (price + speed), Nitro (fastest), Floor (cheapest), or Exacto (highest tool-calling accuracy).
The average price customers actually pay for this model, next to the prices providers post. Caching and discounts mean the price actually paid is often well below the listed one.
Throughput is how fast the model writes (tokens per second — higher is better). Latency is total round-trip time (lower is better). TTFT is time-to-first-token — how long before you see anything appear (lower is better).
Uptime is the percentage of the past 3 days that at least one provider was responding to requests. Availability is the percentage of time that inference was successfully served. OpenRouter continuously monitors and uses the next-best provider when one returns an error.
Scores on standardized evaluations. Higher percentages are better — and rank percentile shows where this model lands among all models on OpenRouter.
| Source | Benchmark | Score |
|---|---|---|
| Artificial Analysis | GPT-4.1 nano Coding Index | 11.1 |
| Artificial Analysis | GPT-4.1 nano GPQA Diamond | 51.2% |
| Artificial Analysis | GPT-4.1 nano HLE | 3.8% |
| Artificial Analysis | GPT-4.1 nano IFBench | 32.0% |
| Artificial Analysis | GPT-4.1 nano τ²-Bench Telecom | 17.3% |
| Artificial Analysis | GPT-4.1 nano AA-LCR | 20.3% |
| Artificial Analysis | GPT-4.1 nano GDPval-AA | 0.0% |
| Artificial Analysis | GPT-4.1 nano CritPt | 0.0% |
| Artificial Analysis | GPT-4.1 nano Terminal-Bench Hard | 3.8% |
| Artificial Analysis | GPT-4.1 nano AA-Omniscience Accuracy | 13.7% |
| Artificial Analysis | GPT-4.1 nano AA-Omniscience Non-Hallucination Rate | 17.4% |
| Design Arena | GPT-4.1 nano Models Arena 3D Elo | 953 |
| Design Arena | GPT-4.1 nano Models Arena Code Categories Elo | 978 |
| Design Arena | GPT-4.1 nano Models Arena Data Visualization Elo | 905 |
| Design Arena | GPT-4.1 nano Models Arena Game Development Elo | 994 |
| Design Arena | GPT-4.1 nano Models Arena UI Component Elo | 931 |
| Design Arena | GPT-4.1 nano Models Arena Website Elo | 985 |
Public apps that send the most traffic to this model. Good signal for what real production workloads look like — and a hint at which use cases this model is best suited for.
Token volume and request traffic to this model over time.
Drop-in code to call this model. OpenRouter's API is OpenAI-compatible — most SDKs work by just swapping the base URL. The only thing that changes between models is the model slug below.
| $0.10 | $0.40 | $0.03 | 1.41s | 11 tps | ||
| $0.10 | $0.40 | $0.025 | 0.92s | 76 tps | ||
Not used in Standard routing:Why these endpoints are not used | ||||||
 | $0.11 | $0.44 | $0.033 | -- | -- | |
Throughput
76tok/s
P50, best across providers
Latency
0.92s
P50, best provider
100.00%
99.99%
When an error occurs in an upstream provider, we can recover by routing to another healthy provider, if your request filters allow it. You can access per-provider uptime data programmatically through the Endpoints API. Learn more about our load balancing and customization options.
For tasks that demand low latency, GPT‑4.1 nano is the fastest and cheapest model in the GPT-4.1 series. It delivers exceptional performance at a small size with its 1 million token context window, and scores 80.1% on MMLU, 50.3% on GPQA, and 9.8% on Aider polyglot coding – even higher than GPT‑4o mini. It’s ideal for tasks like classification or autocompletion.
GPT-4.1 Nano costs $0.10/M input tokens and $0.40/M output tokens, with separate rates for Cache Read at $0.03/M tokens.
GPT-4.1 Nano has a 1,047,576 token context window.
Yes. GPT-4.1 Nano accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
GPT-4.1 Nano accepts images, text, and files such as PDFs as input and returns text.
GPT-4.1 Nano is served by 2 providers on OpenRouter: Azure and OpenAI. Requests are routed to the best available provider, with automatic failover to the others, and you can pin or exclude providers with provider routing.
GPT-4.1 Nano was released on April 14, 2025. Its knowledge cutoff is June 30, 2024.
