# Claude Opus 4.8

> Anthropic’s most capable model for complex reasoning and long-form analysis.

## Overview

- **Model ID**: `claude-opus-4-8`
- **Workflow**: `claude/v1/messages`
- **Provider**: Anthropic
- **Category**: Image → Text (text)
- **Output**: Generated text.
- **Tags**: premium, vision
- **Unified endpoint**: `POST https://api.picsart.com/workflows/v1/models/execute` (model id in the body)
- **Compatibility endpoint**: `POST https://api.picsart.com/workflows/claude/v1/messages/execute` (vendor's own parameters)
- **Execution**: sync (`/execute`) returns the result in one call; async is also available
- **Model page**: https://picsart.com/api-platform/models/claude-opus-4-8

## Pricing

Rates are in USD. Every row below is an exact rate for the settings it names — the columns are the settings that change the price for this model. Every response reports what was actually charged in `response.usage.credits`.

| Tier | Rate |
| --- | --- |
| Cached input | $0.55 / 1M cached input tokens |
| Input | $5.50 / 1M input tokens |
| Cache write 5m | $6.875 / 1M cache write 5m tokens |
| Cache write 1h | $11.00 / 1M cache write 1h tokens |
| Output text | $27.50 / 1M output text tokens |

Published rates. Volume pricing, invoicing and annual agreements: https://picsart.com/api-platform/contact-sales

## Authentication

The SDK and the REST API authenticate with a Picsart API key sent as a bearer token. Create a key in the console at https://picsart.com/api-platform/dashboard/api-keys, keep it in a server-side environment variable, and never ship it in browser code or commit it.

```bash
export PICSART_API_KEY="YOUR_KEY"
```

```http
Authorization: Bearer $PICSART_API_KEY
```

The CLI and the MCP server do not use an API key: `gen-ai login` signs in through the browser once per machine and stores a token locally.

## Interfaces

- **SDK** — `@picsart/ai-sdk` (TypeScript, Node 20+, ESM). Type-safe `ai.generate()` / `ai.generateText()` with per-model parameter types; polls long jobs for you.
- **REST** — the API gateway at `https://api.picsart.com`. Sync `/execute`, or async `/submit` + `/{taskId}/result`.
- **CLI** — `@picsart/gen-ai` (`gen-ai`). Generate from a terminal, quote cost before spending, emit JSON for pipelines.
- **MCP** — the same models inside Claude Code, Cursor, Windsurf and other MCP hosts. Setup: https://picsart.com/gen-ai-mcp/

```bash
npm install @picsart/ai-sdk      # SDK
npm install -g @picsart/gen-ai   # CLI (also drives the MCP server)
```

## Unified API (recommended)

One parameter set, used by the SDK and by REST alike: the names below are what `ai.generate('claude-opus-4-8', …)` takes, and what goes inside the inner `params` of a REST call. Over REST the model id travels in the body, not the path — one workflow, `v1/models`, fronts the whole catalogue.

### Execution modes

- **Sync** — `POST /workflows/v1/models/execute` returns the result in one call. Only for jobs that finish within ~20 seconds; otherwise it times out.
- **Async** — `POST /workflows/v1/models/submit` returns a task id; poll `GET /workflows/v1/models/{taskId}/result` until `response.status` is `COMPLETED` or `FAILED`. Task statuses: `ACCEPTED` · `IN_PROGRESS` · `COMPLETED` · `FAILED`.
- **SDK** — `ai.generate()` always awaits the finished result; polling is hidden.

### Input parameters

- **`prompt`** (`string`, _required_): Prompt
- **`imageUrls`** (`list<string>`, _optional_): Images
  - Accepts: image URLs (max 8)

### Input JSON Schema

The inner `params` object. Over REST it is wrapped as `{ "params": { "model": "<id>", "params": <this> } }`; the SDK takes it directly.

```json
{
  "type": "object",
  "required": [
    "prompt"
  ],
  "properties": {
    "prompt": {
      "type": "string",
      "description": "Prompt"
    },
    "imageUrls": {
      "type": "array",
      "items": {
        "type": "string",
        "format": "uri",
        "description": "image URL"
      },
      "maxItems": 8,
      "description": "Images"
    }
  },
  "additionalProperties": false
}
```

### Required parameters example

The required parameters plus the defaults that shape the output, as a REST body:

```json
{
  "params": {
    "model": "claude-opus-4-8",
    "params": {
      "prompt": "A serene mountain lake at golden hour, ultra detailed"
    }
  }
}
```

### Response

The body is the gateway's status envelope: the output is at `response.result`, next to the task id, its status and the credits charged. Text models return `text`.

```json
{
  "status": "success",
  "response": {
    "id": "task_abc123",
    "status": "COMPLETED",
    "result": {
      "text": "…generated text…",
      "model": "claude-opus-4-8"
    }
  }
}
```

What `ai.generate()` / `ai.generateText()` resolve to (the envelope already unwrapped):

```json
{
  "text": "…generated text…",
  "model": "claude-opus-4-8"
}
```

### Usage examples

#### TypeScript (SDK)

```typescript
import { createClient } from '@picsart/ai-sdk';

const ai = createClient({
  apiKey: process.env.PICSART_API_KEY,
  apiUrl: 'https://api.picsart.com',
});

const { text } = await ai.generateText('claude-opus-4-8', {
  prompt: "A serene mountain lake at golden hour, ultra detailed"
});

console.log(text);
```

#### cURL (sync)

```bash
curl -X POST 'https://api.picsart.com/workflows/v1/models/execute' \
  -H 'Authorization: Bearer $PICSART_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
  "params": {
    "model": "claude-opus-4-8",
    "params": {
      "prompt": "A serene mountain lake at golden hour, ultra detailed"
    }
  }
}'
```

#### CLI

```bash
gen-ai login  # once per machine
gen-ai models info claude-opus-4-8  # this model's parameters
gen-ai pricing claude-opus-4-8  # quote before spending credits
gen-ai generate -m claude-opus-4-8 \
  -p "A serene mountain lake at golden hour"
```

## Compatibility API (vendor parameters)

Alongside the unified API, the `claude/v1/messages` workflow takes Anthropic's own parameters exactly as the vendor defines them — nothing renamed or reshaped. It is the shortest path onto Picsart for request bodies already written against the vendor; the trade is that a request is written for one vendor, and the result comes back in the vendor's own shape. Same authentication.

- **Endpoint**: `POST https://api.picsart.com/workflows/claude/v1/messages/execute` (sync) · `POST …/submit` + `GET …/{taskId}/result` (async)
- **SDK**: `ai.apis.run('claude/v1/messages', params, { mode })` — params forwarded verbatim

### Vendor parameters

| Parameter | Type | Required | Default | Details |
| --- | --- | --- | --- | --- |
| `model` | `string` | yes | — | Claude model to use · Options: `claude-opus-4-8`, `claude-opus-5`, `claude-opus-5-5`, `claude-sonnet-4-5`, `claude-sonnet-4-6`, `claude-sonnet-5`, `claude-sonnet-5-5`, `claude-haiku-4-5`, `claude-haiku-5-5`, `claude-fable-5`, `claude-fable-5-1` |
| `messages` | `ClaudeMessageParam[]` | yes | — | Conversation messages |
| └ `role` | `string` | yes | — | Message role · e.g. `user` |
| └ `content` | `ClaudeContentBlock[]` | yes | — | Message content: a string or an array of content blocks. Passed through to the provider as-is. |
| └   `type` | `string` | yes | — | Content block type · Options: `text`, `tool_use`, `tool_result`, `image`, `thinking`, `file`, `document` |
| └   `text` | `string` | no | — | Text content |
| └   `cache_control` | `object` | no | — | Cache control |
| └   `id` | `string` | no | — | Unique ID for the tool call |
| └   `name` | `string` | no | — | Tool name |
| └   `input` | `string \| number \| integer \| boolean \| array \| object` | no | — | JSON input for the tool |
| └   `tool_use_id` | `string` | no | — | The corresponding tool_use id |
| └   `content` | `ClaudeToolResultContentText[]` | no | — | Tool result content |
| └   `thinking` | `string` | no | — | Internal thinking text |
| └   `signature` | `string` | no | — | Signature is required when echoing thinking back |
| └   `source` | `object` | no | — | Content source |
| `max_tokens` | `number` | no | — | Maximum number of tokens to generate · e.g. `1024` |
| `temperature` | `number` | no | `1` | Sampling temperature, between 0 and 2 · 0–2 |
| `system` | `string \| number \| integer \| boolean \| array \| object` | no | — | System prompt as a string or array of syste blocks |
| `tools` | `ClaudeToolSchema[]` | no | — | List of tools available to the model. Passed through to the provider as-is. |
| └ `name` | `string` | yes | — | Tool name |
| └ `description` | `string` | no | — | Short description of the tool |
| └ `input_schema` | `string \| number \| integer \| boolean \| array \| object` | yes | — | JSON Schema for the tool input |
| └ `type` | `string` | yes | `custom` | — |
| `tool_choice` | `ClaudeToolChoiceToolParam \| string` | no | — | Controls which (if any) tool is called by the model. "auto" lets the model decide; "any" forces a tool call; { type: "tool", name } forces a specific tool. |
| `service_tier` | `string` | no | — | Specifies the processing type used for serving the request. Passed through to the provider as-is. · Options: `auto`, `flex`, `priority`, `default` |
| `context_management` | `object` | no | — | Context management options. Passed through to the provider as-is. |
| └ `edits` | `ContextManagementEdit[]` | yes | — | List of context management edits to apply |
| └   `type` | `string` | yes | `clear_tool_uses_20250919` | clear_thinking_20251015  or clear_tool_uses_20250919 |
| └   `trigger` | `object` | no | — | When to trigger this edit |
| └   `keep` | `ContextCriterion \| string` | no | — | What to keep when applying the edit: a criterion object or a keyword such as "all". |
| └   `clear_at_least` | `object` | no | — | Minimum to clear when applying the edit |
| └   `exclude_tools` | `string[]` | no | — | Exclude these tool names from clearing |
| └   `clear_tool_inputs` | `boolean` | no | `false` | Whether to clear the tool uses |
| `options` | `object` | no | — | — |
| └ `safety_checks` | `object` | no | — | Safety check settings |
| └   `enabled` | `boolean` | no | — | Whether to run content moderation. Defaults to true. |
| └ `drive` | `object` | no | — | Save result to Picsart Drive |
| └   `name` | `string` | yes | — | File name in Picsart Drive |
| └   `attributes` | `object` | no | — | Custom attributes to attach to the file |
| └   `folder` | `object` | no | — | Target folder in Picsart Drive |
| └ `inputs_transformation` | `object` | no | — | Input transformation settings |
| └   `downscale_oversized_images` | `boolean` | no | — | Whether to downscale oversized input images. Defaults to false. |

### Example request body

```json
{
  "params": {
    "model": "claude-opus-4-8",
    "messages": [
      {
        "role": "user",
        "content": []
      }
    ]
  }
}
```

### Usage examples

#### TypeScript (`ai.apis.run`)

```typescript
import { createClient, ApiRunMode } from '@picsart/ai-sdk';

const ai = createClient({
  apiKey: process.env.PICSART_API_KEY,
  apiUrl: 'https://api.picsart.com',
});

// Calls the 'claude/v1/messages' workflow directly — params are sent as-is.
const { result, usage } = await ai.apis.run('claude/v1/messages', {
  model: "claude-opus-4-8",
  messages: [
    {
      role: "user",
      content: []
    }
  ]
}, {
  mode: ApiRunMode.SYNC,
});

console.log(result); // workflow-specific output
console.log(usage?.credits); // amount charged, in $
```

#### cURL (sync)

```bash
curl -X POST 'https://api.picsart.com/workflows/claude/v1/messages/execute' \
  -H 'Authorization: Bearer $PICSART_API_KEY' \
  -H 'Content-Type: application/json' \
  -d '{
  "params": {
    "model": "claude-opus-4-8",
    "messages": [
      {
        "role": "user",
        "content": []
      }
    ]
  }
}'
```

### Response

Over HTTP the vendor-shaped result arrives inside the gateway envelope at `response.result`:

```json
{
  "status": "success",
  "response": {}
}
```

`ai.apis.run()` unwraps that envelope and resolves to `{ result, usage }`:

```json
{
  "result": {}
}
```

## Additional resources

- Model page: https://picsart.com/api-platform/models/claude-opus-4-8
- Try it: https://picsart.com/api-platform/models/claude-opus-4-8/playground
- SDK: https://picsart.com/api-platform/models/claude-opus-4-8/sdk
- REST API (unified): https://picsart.com/api-platform/models/claude-opus-4-8/api
- Compatibility API (vendor parameters): https://picsart.com/api-platform/models/claude-opus-4-8/advanced
- Pricing: https://picsart.com/api-platform/models/claude-opus-4-8/pricing
- This document: https://picsart.com/api-platform/models/claude-opus-4-8/llms.txt
- All models: https://picsart.com/api-platform/models
- Docs — quickstart https://picsart.com/api-platform/docs/quickstart, authentication https://picsart.com/api-platform/docs/authentication, SDK https://picsart.com/api-platform/docs/sdk, REST https://picsart.com/api-platform/docs/api-reference, CLI https://picsart.com/api-platform/docs/cli, MCP https://picsart.com/api-platform/docs/mcp
- CLI reference: https://picsart.github.io/picsart-mcp-cli-docs/guide/introduction