> ## Documentation Index
> Fetch the complete documentation index at: https://docs.apiyi.com/llms.txt
> Use this file to discover all available pages before exploring further.
# GPT-5.4 Mini & Nano Launch: Lightweight, Cost-Effective Models Built for Scale
> OpenAI released GPT-5.4 Mini and Nano on March 17, 2026. Mini runs 2x faster than GPT-5 Mini, Nano costs just $0.20/M input tokens. Now available on APIYI at official pricing with ~10% deposit bonus.
## Key Highlights
* **Most Capable Small Models**: GPT-5.4 Mini significantly improves over GPT-5 Mini across coding, reasoning, multimodal understanding, and tool use — over 2x faster
* **Ultra Cost-Effective**: GPT-5.4 Nano at just \$0.20/M input tokens and \$1.25/M output tokens, the cheapest GPT-5.4 family model
* **Near-Flagship Performance**: Mini scores 54.4% on SWE-Bench Pro and 72.1% on OSWorld-Verified, approaching full GPT-5.4
* **400K Context Window**: Consistent with the GPT-5 family
* **Full Capabilities**: Text & image input, tool calling, web search, computer use — all supported
## Background
On March 17, 2026, OpenAI officially released GPT-5.4 Mini and GPT-5.4 Nano, calling them "the most capable small models yet." Following the GPT-5.4 flagship launch in early March, this release brings 5.4-level capabilities to lightweight, cost-efficient models.
GPT-5.4 Mini targets developers who need high performance on a budget, delivering near-flagship quality at a fraction of the cost. GPT-5.4 Nano is purpose-built for high-throughput, low-cost workloads like classification, data extraction, ranking, and coding subagents.
APIYI has immediately launched both models at pricing identical to OpenAI's official rates, with official API proxy and \~10% deposit bonus.
## Detailed Analysis
### Core Features
GPT-5.4 Mini runs more than 2x faster than GPT-5 Mini with significantly lower latency
Nano can describe 76,000 photos for approximately \$52
Mini scores 54.4% on SWE-Bench Pro (vs GPT-5.4's 57.7%)
Image understanding, tool calling, web search, computer use — all included
### GPT-5.4 Mini Benchmarks
GPT-5.4 Mini dominates across evaluations compared to GPT-5 Mini:
| Benchmark | GPT-5.4 Mini | GPT-5.4 (Full) | GPT-5 Mini | Description |
| - | - | - | - | - |
| **SWE-Bench Pro** | **54.4%** | 57.7% | 45.7% | Real-world software engineering |
| **OSWorld-Verified** | **72.1%** | 75.0% | 42.0% | Computer use benchmark |
| **Toolathlon** | **42.9%** | — | 26.9% | Tool calling evaluation |
| **GPQA Diamond** | **88.0%** | — | 81.6% | Research-level science reasoning |
| **Tau2-Bench** | **93.4%** | — | 74.1% | Tool calling benchmark |
| **MCP Atlas** | **57.7%** | — | 47.6% | MCP protocol evaluation |
| **Terminal-Bench 2.0** | **60.0%** | — | — | Terminal operation evaluation |
GPT-5.4 Mini scores 72.1% on OSWorld-Verified, a 71.7% improvement over GPT-5 Mini's 42.0%, approaching the flagship GPT-5.4's 75.0%.
### GPT-5.4 Nano Performance
Nano is the smallest, cheapest GPT-5.4 variant, designed for speed and cost-first scenarios:
| Benchmark | GPT-5.4 Nano | GPT-5.4 Mini |
| - | - | - |
| **Terminal-Bench 2.0** | 46.3% | 60.0% |
| **OSWorld-Verified** | 39.0% | 72.1% |
OpenAI recommends Nano for classification, data extraction, ranking, and coding subagents handling simpler supporting tasks.
### Adjustable Reasoning Effort
Both models support variable reasoning effort levels for flexible cost-performance tradeoffs:
* `none`: No reasoning, fastest response
* `low` / `medium` / `high`: Increasing reasoning depth
* `xhigh`: Maximum reasoning effort
Notably, GPT-5.4 Nano at maximum reasoning effort outperforms the previous GPT-5 Mini, achieving last-generation mid-tier capabilities at a fraction of the cost.
## Practical Applications
### Recommended Use Cases
**GPT-5.4 Mini:**
1. **Coding Assistant**: 54.4% SWE-Bench Pro, ideal for code generation, review, and debugging
2. **Autonomous Agents**: 72.1% OSWorld, supports computer use and multi-step workflows
3. **Daily Chat**: Default thinking model for ChatGPT Free and Go users
4. **OpenClaw & Similar**: High performance at low cost, perfect for batch intelligent tasks
**GPT-5.4 Nano:**
1. **Data Pipelines**: Classification, extraction, ranking at high throughput
2. **Coding Subagents**: Handle simpler supporting tasks in agent architectures
3. **Batch Image Understanding**: \~\$52 for 76,000 photo descriptions
4. **Real-time Classification**: Ultra-low latency, ultra-low cost
### Code Examples
#### Using GPT-5.4 Mini
```python theme={null}
from openai import OpenAI
client = OpenAI(
api_key="your-apiyi-key",
base_url="https://api.apiyi.com/v1"
)
# GPT-5.4 Mini - Cost-effective coding assistant
response = client.chat.completions.create(
model="gpt-5.4-mini",
messages=[
{
"role": "user",
"content": "Refactor this code to extract common logic and add error handling..."
}
],
max_tokens=4096
)
print(response.choices[0].message.content)
```
#### Using GPT-5.4 Nano for Batch Processing
```python theme={null}
# GPT-5.4 Nano - Ultra cost-effective batch tasks
response = client.chat.completions.create(
model="gpt-5.4-nano",
messages=[
{
"role": "system",
"content": "You are a text classifier. Classify user input as: positive, negative, or neutral. Output only the classification."
},
{
"role": "user",
"content": "This product is excellent quality, will definitely buy again!"
}
],
max_tokens=10
)
print(response.choices[0].message.content)
```
## Pricing & Availability
### Pricing
| Tier | GPT-5.4 Nano | GPT-5.4 Mini | GPT-5.4 (Full) |
| - | - | - | - |
| **Input** | \$0.20 / M tokens | \$0.75 / M tokens | \$2.50 / M tokens |
| **Cached Input** | \$0.02 / M tokens | \$0.075 / M tokens | — |
| **Output** | \$1.25 / M tokens | \$4.50 / M tokens | \$15.00 / M tokens |
GPT-5.4 Mini costs just 30% of the flagship GPT-5.4 with near-flagship performance. Nano is only 8% of flagship cost — ultimate value.
**Competitive Comparison**:
| Model | Input Price | Output Price | Positioning |
| - | - | - | - |
| **GPT-5.4 Nano** | **\$0.20** | **\$1.25** | Ultra-lightweight |
| **GPT-5.4 Mini** | **\$0.75** | **\$4.50** | Cost-effective |
| Gemini 3.1 Flash Lite | \$0.25 | \$0.50 | Lightweight fast |
| Claude Haiku 4.5 | \$0.80 | \$4.00 | Fast response |
| GPT-5.2 mini | \$0.30 | \$1.80 | Previous-gen small model |
### Promotions
APIYI offers deposit bonuses — official API proxy with \~10% bonus on deposits. Pricing matches official rates, with effective discounts through deposit bonuses.
### Available Models
| Model Name | Variant | Context | Description |
| - | - | - | - |
| `gpt-5.4-mini` | Mini | 400K | Cost-effective, near-flagship performance |
| `gpt-5.4-nano` | Nano | 400K | Ultra-low cost, high-throughput scenarios |
### How to Access
**APIYI Platform**:
* Website: `apiyi.com`
* API Endpoint: `https://api.apiyi.com/v1`
* OpenAI-compatible format, works with all OpenAI SDKs
* Official API proxy, stable and reliable
## Summary & Recommendations
GPT-5.4 Mini and Nano bring flagship-level capabilities to developers and applications at dramatically lower costs.
**Key Advantages**:
* **Mini**: Flagship-tier capability at 30% cost, 2x speed
* **Nano**: Ultra-low pricing at 8% of flagship cost, built for scale
**Selection Guide**:
1. **Near-flagship performance needed**: Choose GPT-5.4 Mini (SWE-Bench Pro 54.4%, OSWorld 72.1%)
2. **High-throughput batch processing**: Choose GPT-5.4 Nano (\$0.20/M input tokens)
3. **Agent architecture subtasks**: Nano as execution layer, Mini as decision layer
4. **OpenClaw & similar**: Both Mini and Nano work well — choose based on cost-performance needs
5. **Cost-insensitive professional tasks**: Still recommend GPT-5.4 flagship or Pro
APIYI has launched both GPT-5.4 Mini and Nano at official pricing with \~10% deposit bonus. Experience OpenAI's most capable lightweight models today!
Sources: OpenAI official blog (March 17, 2026), 9to5Mac, The New Stack, Simon Willison's Weblog, and other authoritative sources. Data retrieved: March 18, 2026.
This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.