> ## Documentation Index
> Fetch the complete documentation index at: https://docs.wolffi.sh/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenAI

> Set up OpenAI (GPT) — general-purpose tasks, broad knowledge, fast responses

# OpenAI (GPT)

```
POST https://api.openai.com/v1/chat/completions
```

Uses SSE streaming. Tool calls arrive as `function_call` objects.

Best for: General-purpose tasks, broad knowledge, fast responses.

## Getting an API Key

1. Go to [platform.openai.com](https://platform.openai.com)
2. Sign up or log in
3. Navigate to **API Keys** and create a new key
4. Paste it into Wolffish → Settings → Models → OpenAI

## Models

### GPT-5 family (reasoning)

| Model        | Context | Modes          | Input / Output (per MTok) | Notes                                              |
| ------------ | ------- | -------------- | ------------------------- | -------------------------------------------------- |
| **gpt-5.5**  | 1M      | Off, High, Max | $5.00 / $30.00            | Flagship. Frontier reasoning. Cached: \$0.50/MTok. |
| gpt-5.4      | 1M      | Off, High, Max | $2.50 / $15.00            | Cached: \$0.25/MTok.                               |
| gpt-5.4-mini | 1M      | Off, High, Max | $0.75 / $4.50             | Fast reasoning. Cached: \$0.08/MTok.               |
| gpt-5.4-nano | 1M      | Off, High, Max | $0.20 / $1.25             | Ultra-cheap reasoning. Cached: \$0.02/MTok.        |
| gpt-5.2      | 1M      | Off, High, Max | — / —                     | Pricing TBD.                                       |
| gpt-5.1      | 1M      | Off, High, Max | — / —                     | Pricing TBD.                                       |
| gpt-5        | 1M      | Off, High, Max | $2.50 / $10.00            | Cached: \$1.25/MTok.                               |
| gpt-5-mini   | 1M      | Off, High, Max | $0.25 / $2.00             | Fast reasoning. Cached: \$0.03/MTok.               |
| gpt-5-nano   | 1M      | Off, High, Max | $0.05 / $0.40             | Fast reasoning. Cached: \$0.01/MTok.               |

### o-series (reasoning)

| Model   | Context | Modes          | Input / Output (per MTok) | Notes                                |
| ------- | ------- | -------------- | ------------------------- | ------------------------------------ |
| o3      | 200K    | Off, High, Max | $10.00 / $40.00           | Cached: \$5.00/MTok.                 |
| o4-mini | 200K    | Off, High, Max | $1.10 / $4.40             | Fast reasoning. Cached: \$0.55/MTok. |
| o3-mini | 200K    | Off, High, Max | $1.10 / $4.40             | Fast reasoning. Cached: \$0.55/MTok. |
| o1      | 200K    | Off, High, Max | $15.00 / $60.00           | Cached: \$7.50/MTok.                 |

### GPT-4 family (non-reasoning)

| Model        | Context | Modes | Input / Output (per MTok) | Notes                      |
| ------------ | ------- | ----- | ------------------------- | -------------------------- |
| gpt-4.1      | 1M      | —     | $2.00 / $8.00             | Cached: \$0.50/MTok.       |
| gpt-4.1-mini | 1M      | —     | $0.40 / $1.60             | Fast. Cached: \$0.10/MTok. |
| gpt-4.1-nano | 1M      | —     | $0.10 / $0.40             | Fast. Cached: \$0.03/MTok. |
| gpt-4o       | 128K    | —     | $2.50 / $10.00            | Cached: \$1.25/MTok.       |
| gpt-4o-mini  | 128K    | —     | $0.15 / $0.60             | Fast. Cached: \$0.08/MTok. |
| gpt-4-turbo  | 128K    | —     | $10.00 / $30.00           |                            |
| gpt-4        | 8K      | —     | $30.00 / $60.00           |                            |

## Reasoning modes

The **brain icon** next to the message box controls how this model reasons. Click it to cycle through the modes the selected model supports. Two separate ideas combine here:

### Thinking — *whether* the model reasons

* **Off** — the model answers immediately. Fastest and cheapest; ideal for simple, direct tasks.
* **On** — the model first works through the problem in a dedicated reasoning pass before replying. Slower and uses more tokens, but markedly more accurate on multi-step, logical, or ambiguous tasks.

### Effort — *how hard* it thinks

Only effort-capable models expose this; it applies once thinking is on.

* **High** — standard reasoning depth. The right default for most agentic work.
* **Max** — the model reasons longer and deeper for the hardest problems. More tokens and latency in exchange for higher quality on complex work.

### Button states

| State | Colour | Meaning                         |
| ----- | ------ | ------------------------------- |
| Off   | gray   | Thinking off — direct answer    |
| On    | blue   | Thinking on — no effort control |
| High  | purple | Thinking on, standard effort    |
| Max   | orange | Thinking on, maximum effort     |

Each model shows only the states it genuinely supports. If a model always reasons (can't be turned off) or has no effort control, the button reflects that and locks where there's nothing to change. Wolffish remembers your choice per model.

**On OpenAI:** GPT-5 and o-series models support Off / High / Max (Max maps to OpenAI's `xhigh` effort). Note: OpenAI's chat API can't combine reasoning effort with tool calls, so during tool-using turns Wolffish drops the effort and the model reasons at its default.