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

# GPT-5.5

> OpenAI's current production flagship — 128K context, native reasoning, the model to pick for any new project that doesn't have a benchmark for an older version.

**Vendor:** OpenAI
**Model ID:** `gpt-5-5`
**Capability:** 128K context · tool use · vision · streaming · structured output · reasoning\_effort
**Pricing:** per-token, flagship tier ([live rate](https://bytespike.ai/pricing#text))

GPT-5.5 is the current OpenAI flagship — the default for any new
project on the platform. It's the model that put native reasoning
into the standard chat completions shape: same request body, same
response shape, with `reasoning_effort` as an optional dial. For most
production work the default `"medium"` setting is right; lift to
`"high"` only on hard problems where Sonnet or 5.4-pro have left
quality on the table.

## Request

```bash theme={null}
curl https://llm.bytespike.ai/v1/chat/completions \
  -H "Authorization: Bearer $BYTESPIKE_API_KEY" \
  -H "content-type: application/json" \
  -d '{
    "model": "gpt-5-5",
    "messages": [{"role": "user", "content": "Design a schema for a multi-tenant audit log."}]
  }'
```

### Body parameters

| Field              | Type    | Required | Default    | Notes                                                       |
| ------------------ | ------- | -------- | ---------- | ----------------------------------------------------------- |
| `model`            | string  | yes      | —          | `gpt-5-5`                                                   |
| `messages`         | array   | yes      | —          | —                                                           |
| `reasoning_effort` | string  | no       | `"medium"` | `"low"` / `"medium"` / `"high"`.                            |
| `max_tokens`       | integer | no       | model max  | Max: 32768.                                                 |
| `tools`            | array   | no       | —          | Parallel function calling.                                  |
| `response_format`  | object  | no       | —          | JSON mode + structured output (recommended for production). |
| `web_search`       | object  | no       | —          | Built-in web search tool — billed per use.                  |
| `stream`           | boolean | no       | false      | SSE streaming.                                              |

## Response

```json theme={null}
{
  "id": "chatcmpl-…",
  "object": "chat.completion",
  "model": "gpt-5-5",
  "choices": [{"index": 0, "message": {"role": "assistant", "content": "..."}, "finish_reason": "stop"}],
  "usage": {
    "prompt_tokens": 64,
    "completion_tokens": 1842,
    "reasoning_tokens": 2048,
    "total_tokens": 3954
  }
}
```

`reasoning_tokens` billed at input-token rate.

## Code examples

<CodeGroup>
  ```bash cURL theme={null}
  curl https://llm.bytespike.ai/v1/chat/completions \
    -H "Authorization: Bearer $BYTESPIKE_API_KEY" \
    -H "content-type: application/json" \
    -d '{"model": "gpt-5-5", "messages": [{"role": "user", "content": "Design a schema for a multi-tenant audit log."}]}'
  ```

  ```python Python theme={null}
  from openai import OpenAI

  client = OpenAI(base_url="https://llm.bytespike.ai/v1", api_key="$BYTESPIKE_API_KEY")
  resp = client.chat.completions.create(
      model="gpt-5-5",
      messages=[{"role": "user", "content": "Design a schema for a multi-tenant audit log."}],
  )
  print(resp.choices[0].message.content)
  ```

  ```javascript Node theme={null}
  import OpenAI from "openai"

  const client = new OpenAI({
    baseURL: "https://llm.bytespike.ai/v1",
    apiKey: process.env.BYTESPIKE_API_KEY,
  })
  const resp = await client.chat.completions.create({
    model: "gpt-5-5",
    messages: [
      { role: "user", content: "Design a schema for a multi-tenant audit log." },
    ],
  })
  console.log(resp.choices[0].message.content)
  ```
</CodeGroup>

## Reasoning effort

| Setting    | Use for                                                          |
| ---------- | ---------------------------------------------------------------- |
| `"low"`    | Fast routing / classification with reasoning kept light          |
| `"medium"` | Default — production code gen, content rewriting, agents         |
| `"high"`   | Hard problems where you can wait — proofs, deep refactors, plans |

## Web search

Pass `"web_search": {}` to give the model a built-in web search tool.
The tool is billed per use (see [pricing](https://bytespike.ai/pricing#text)
for current rate). Useful for fact-grounded tasks where the model would
otherwise hallucinate or cite stale information.

## Streaming + caching

`"stream": true` for SSE. With reasoning enabled, expect a longer TTFB.
Automatic prompt caching on stable prefixes — the highest-leverage cost
optimisation on this tier.

## Errors

| Code                        | Trigger  | Billed?         |
| --------------------------- | -------- | --------------- |
| 400 / 401 / 402 / 422 / 429 | Standard | No              |
| 5xx                         | Upstream | No (auto-retry) |

## When to use

* Default starting point for any new project on OpenAI.
* Code generation in an existing codebase, schema / API design.
* Multi-step plans, structured output where mid-tier models drift.
* For mid-tier cost / latency, see [GPT-5.4](/api-reference/text/gpt-5-4).

## Limits

| Limit                      | Value          |
| -------------------------- | -------------- |
| Context window             | 128K tokens    |
| Max output                 | 32768 tokens   |
| Supports tool use          | Yes (parallel) |
| Supports vision            | Yes            |
| Supports streaming         | Yes            |
| Supports prompt caching    | Automatic      |
| Supports reasoning\_effort | Yes            |
| Supports web search tool   | Yes            |