Pointing

from vlmrun.client import VLMRun

# Initialize the VLMRun client
client = VLMRun(api_key="<VLMRUN_API_KEY>")

# Predict the keypoints in the image
response = client.agent.completions.create(
    model="vlmrun-orion-1:auto",
    messages=[
        {
          "role": "user",
          "content": [
            {"type": "text", "text": "Point to all the cars and doors in this image"},
            {"type": "image_url", "image_url": {"url": "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/image.caption/car.jpg", "detail": "auto"}}
          ]
        }
    ],
)

print(response.choices[0].message.content)

from vlmrun.client import VLMRun
from pydantic import BaseModel, Field

# Define the response schema
class KeyPoint(BaseModel):
  xy: tuple[float, float] = Field(..., description="Normalized keypoint coordinates [x, y]")
  label: str = Field(..., description="Label of the keypoint")

class Keypoints(BaseModel):
  keypoints: list[KeyPoint] = Field(..., description="List of keypoints")

# Initialize the VLMRun client
client = VLMRun(api_key="<VLMRUN_API_KEY>")

# Predict the keypoints with structured output
response = client.agent.completions.create(
    model="vlmrun-orion-1:auto",
    messages=[
        {
          "role": "user",
          "content": [
            {"type": "text", "text": "Point to all the cars and doors in this image"},
            {"type": "image_url", "image_url": {"url": "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/image.caption/car.jpg", "detail": "auto"}}
          ]
        }
    ],
    response_format={"type": "json_schema", "schema": Keypoints.model_json_schema()},
)

# Validate the response
result = Keypoints.model_validate_json(response.choices[0].message.content)
# >>> Keypoints(keypoints=[KeyPoint(xy=(0.5, 0.5), label='car'), ...])

import { VlmRun } from "vlmrun";

const client = new VlmRun({
  apiKey: "<VLMRUN_API_KEY>",
  baseURL: "https://api.vlm.run/v1"
});

const response = await client.agent.completions.create({
  model: "vlmrun-orion-1:auto",
  messages: [
    {
      role: "user",
      content: [
        { type: "text", text: "Point to all the cars and doors in this image" },
        { type: "image_url", image_url: { url: "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/image.caption/car.jpg", detail: "auto" } }
      ]
    }
  ]
});

console.log(response.choices[0].message.content);

import { VlmRun } from "vlmrun";
import { z } from "zod";
import { zodToJsonSchema } from "zod-to-json-schema";

// Define the response schema with Zod
const KeypointsSchema = z.object({
  keypoints: z.array(z.object({
    xy: z.array(z.number()).describe("Normalized keypoint coordinates [x, y]"),
    label: z.string().describe("Label of the keypoint")
  })).describe("List of keypoints")
});

// Initialize the VLMRun client
const client = new VlmRun({
  apiKey: "<VLMRUN_API_KEY>",
  baseURL: "https://api.vlm.run/v1"
});

// Predict the keypoints with structured output
const response = await client.agent.completions.create({
  model: "vlmrun-orion-1:auto",
  messages: [
    {
      role: "user",
      content: [
        { type: "text", text: "Point to all the cars and doors in this image" },
        { type: "image_url", image_url: { url: "https://storage.googleapis.com/vlm-data-public-prod/hub/examples/image.caption/car.jpg", detail: "auto" } }
      ]
    }
  ],
  response_format: {
    type: "json_schema",
    schema: zodToJsonSchema(KeypointsSchema)
  }
});

const result = KeypointsSchema.parse(JSON.parse(response.choices[0].message.content));