deepseek-ai / deepseek-vl-7b-base

{
  "image": "https://replicate.delivery/pbxt/KYKAXfcjSZ7uEkwQoYG4SFXRJpbGcdKhWow4Ul0y4Qkjw6wW/training_pipelines.png",
  "prompt": "Describe each stage of this image",
  "max_new_tokens": 512
}

Install Replicate’s Node.js client library:

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
  {
    input: {
      image: "https://replicate.delivery/pbxt/KYKAXfcjSZ7uEkwQoYG4SFXRJpbGcdKhWow4Ul0y4Qkjw6wW/training_pipelines.png",
      prompt: "Describe each stage of this image",
      max_new_tokens: 512
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Import the client:

import replicate

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
    input={
        "image": "https://replicate.delivery/pbxt/KYKAXfcjSZ7uEkwQoYG4SFXRJpbGcdKhWow4Ul0y4Qkjw6wW/training_pipelines.png",
        "prompt": "Describe each stage of this image",
        "max_new_tokens": 512
    }
)

# The deepseek-ai/deepseek-vl-7b-base model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/deepseek-ai/deepseek-vl-7b-base/api#output-schema
    print(item, end="")

To learn more, take a look at the guide on getting started with Python.

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
    "input": {
      "image": "https://replicate.delivery/pbxt/KYKAXfcjSZ7uEkwQoYG4SFXRJpbGcdKhWow4Ul0y4Qkjw6wW/training_pipelines.png",
      "prompt": "Describe each stage of this image",
      "max_new_tokens": 512
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

The image depicts a three-stage process for training a vision-language model. 1. Stage 1: Training VL Adapter: In this stage, a vision-language adapter is trained using supervised fine-tuning. The adapter is trained on image-text pairs and pure language sequences. 2. Stage 2: Joint VL Pre-training: In this stage, a joint vision-language model is pre-trained using self-supervised learning. The model is trained on image-text pairs and pure language sequences. 3. Stage 3: Supervised Fine-tuning: In this stage, the model is fine-tuned on supervised tasks using image-text pairs and pure language sequences. The model is trained using a hybrid vision-language adapter, which combines a vision-language adapter with a language model. The model is trained on a variety of tasks, including image captioning, visual question answering, and visual reasoning. The model is able to understand the visual content of an image and generate a natural language description or answer.

{
  "completed_at": "2024-03-11T19:55:28.030312Z",
  "created_at": "2024-03-11T19:54:05.161849Z",
  "data_removed": false,
  "error": null,
  "id": "sghdgetbu4kminwy4vmljtugou",
  "input": {
    "image": "https://replicate.delivery/pbxt/KYKAXfcjSZ7uEkwQoYG4SFXRJpbGcdKhWow4Ul0y4Qkjw6wW/training_pipelines.png",
    "prompt": "Describe each stage of this image",
    "max_new_tokens": 512
  },
  "logs": null,
  "metrics": {
    "predict_time": 8.43218,
    "total_time": 82.868463
  },
  "output": [
    " ",
    "The ",
    "image ",
    "depicts ",
    "a ",
    "",
    "",
    "three-stage ",
    "process ",
    "for ",
    "training ",
    "a ",
    "",
    "",
    "vision-language ",
    "",
    "model.\n",
    "\n",
    "",
    "",
    "1. ",
    "Stage ",
    "",
    "",
    "1: ",
    "Training ",
    "VL ",
    "",
    "Adapter: ",
    "In ",
    "this ",
    "",
    "stage, ",
    "a ",
    "",
    "",
    "vision-language ",
    "adapter ",
    "is ",
    "trained ",
    "using ",
    "supervised ",
    "",
    "",
    "",
    "fine-tuning. ",
    "The ",
    "adapter ",
    "is ",
    "trained ",
    "on ",
    "",
    "",
    "image-text ",
    "pairs ",
    "and ",
    "pure ",
    "language ",
    "",
    "sequences.\n",
    "",
    "",
    "2. ",
    "Stage ",
    "",
    "",
    "2: ",
    "Joint ",
    "VL ",
    "",
    "",
    "",
    "Pre-training: ",
    "In ",
    "this ",
    "",
    "stage, ",
    "a ",
    "joint ",
    "",
    "",
    "vision-language ",
    "model ",
    "is ",
    "",
    "",
    "pre-trained ",
    "using ",
    "",
    "",
    "self-supervised ",
    "",
    "learning. ",
    "The ",
    "model ",
    "is ",
    "trained ",
    "on ",
    "",
    "",
    "image-text ",
    "pairs ",
    "and ",
    "pure ",
    "language ",
    "",
    "sequences.\n",
    "",
    "",
    "3. ",
    "Stage ",
    "",
    "",
    "3: ",
    "",
    "Supervised ",
    "",
    "",
    "",
    "Fine-tuning: ",
    "In ",
    "this ",
    "",
    "stage, ",
    "the ",
    "model ",
    "is ",
    "",
    "",
    "fine-tuned ",
    "on ",
    "supervised ",
    "tasks ",
    "using ",
    "",
    "",
    "image-text ",
    "pairs ",
    "and ",
    "pure ",
    "language ",
    "",
    "sequences.\n",
    "\n",
    "",
    "The ",
    "model ",
    "is ",
    "trained ",
    "using ",
    "a ",
    "hybrid ",
    "",
    "",
    "vision-language ",
    "",
    "adapter, ",
    "which ",
    "combines ",
    "a ",
    "",
    "",
    "vision-language ",
    "adapter ",
    "with ",
    "a ",
    "language ",
    "",
    "model. ",
    "The ",
    "model ",
    "is ",
    "trained ",
    "on ",
    "a ",
    "variety ",
    "of ",
    "",
    "tasks, ",
    "including ",
    "image ",
    "",
    "",
    "captioning, ",
    "visual ",
    "question ",
    "",
    "answering, ",
    "and ",
    "visual ",
    "",
    "reasoning. ",
    "The ",
    "model ",
    "is ",
    "able ",
    "to ",
    "understand ",
    "the ",
    "visual ",
    "content ",
    "of ",
    "an ",
    "image ",
    "and ",
    "generate ",
    "a ",
    "natural ",
    "language ",
    "description ",
    "or ",
    "",
    "",
    "answer."
  ],
  "started_at": "2024-03-11T19:55:19.598132Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://streaming-api.svc.us.c.replicate.net/v1/streams/6duul3ruer2sfbowces56rgdkofnjyeaxi36sx2vggdlhdnxlq7a",
    "get": "https://api.replicate.com/v1/predictions/sghdgetbu4kminwy4vmljtugou",
    "cancel": "https://api.replicate.com/v1/predictions/sghdgetbu4kminwy4vmljtugou/cancel"
  },
  "version": "d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385"
}

Generated in

8.4 seconds

Tweak it Share Report

Prediction

deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385

Model

deepseek-ai/deepseek-vl-7b-base:d1823e6f

xfwnzedbxhtwqftqabcj5aiaf4

Status

Succeeded

Source

Web

Hardware

A40 (Large)

Total duration

83.1s

Created

about 1 year ago

Input

image
prompt: Describe this image
max_new_tokens: 512

{
  "image": "https://replicate.delivery/pbxt/KYKzW8HpnrLwjibeVRBozgVvHKpbtyglrzrixoYJJr2toFzK/monday.jpg",
  "prompt": "Describe this image",
  "max_new_tokens": 512
}

Install Replicate’s Node.js client library:

npm install replicate

Import and set up the client:

import Replicate from "replicate";

const replicate = new Replicate({
  auth: process.env.REPLICATE_API_TOKEN,
});

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

const output = await replicate.run(
  "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
  {
    input: {
      image: "https://replicate.delivery/pbxt/KYKzW8HpnrLwjibeVRBozgVvHKpbtyglrzrixoYJJr2toFzK/monday.jpg",
      prompt: "Describe this image",
      max_new_tokens: 512
    }
  }
);

console.log(output);

To learn more, take a look at the guide on getting started with Node.js.

Install Replicate’s Python client library:

pip install replicate

Import the client:

import replicate

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

output = replicate.run(
    "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
    input={
        "image": "https://replicate.delivery/pbxt/KYKzW8HpnrLwjibeVRBozgVvHKpbtyglrzrixoYJJr2toFzK/monday.jpg",
        "prompt": "Describe this image",
        "max_new_tokens": 512
    }
)

# The deepseek-ai/deepseek-vl-7b-base model can stream output as it's running.
# The predict method returns an iterator, and you can iterate over that output.
for item in output:
    # https://replicate.com/deepseek-ai/deepseek-vl-7b-base/api#output-schema
    print(item, end="")

To learn more, take a look at the guide on getting started with Python.

Run deepseek-ai/deepseek-vl-7b-base using Replicate’s API. Check out the model's schema for an overview of inputs and outputs.

curl -s -X POST \
  -H "Authorization: Bearer $REPLICATE_API_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Prefer: wait" \
  -d $'{
    "version": "deepseek-ai/deepseek-vl-7b-base:d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385",
    "input": {
      "image": "https://replicate.delivery/pbxt/KYKzW8HpnrLwjibeVRBozgVvHKpbtyglrzrixoYJJr2toFzK/monday.jpg",
      "prompt": "Describe this image",
      "max_new_tokens": 512
    }
  }' \
  https://api.replicate.com/v1/predictions

To learn more, take a look at Replicate’s HTTP API reference docs.

Output

The image you've provided is a meme featuring a French Bulldog lying on its side on a blue wooden floor. The dog's head is resting on its paws, and it appears to be sleeping. The meme has a text overlay that reads "MONDAY. JUST...MONDAY." This suggests that the meme is making a humorous comment about the start of the work week. The dog's relaxed posture and the text overlay together create a light-hearted and relatable message.

{
  "completed_at": "2024-03-11T19:58:52.142045Z",
  "created_at": "2024-03-11T19:57:29.036420Z",
  "data_removed": false,
  "error": null,
  "id": "xfwnzedbxhtwqftqabcj5aiaf4",
  "input": {
    "image": "https://replicate.delivery/pbxt/KYKzW8HpnrLwjibeVRBozgVvHKpbtyglrzrixoYJJr2toFzK/monday.jpg",
    "prompt": "Describe this image",
    "max_new_tokens": 512
  },
  "logs": null,
  "metrics": {
    "predict_time": 5.700052,
    "total_time": 83.105625
  },
  "output": [
    " ",
    "\n",
    "\n",
    "",
    "The ",
    "image ",
    "",
    "",
    "you've ",
    "provided ",
    "is ",
    "a ",
    "meme ",
    "featuring ",
    "a ",
    "French ",
    "",
    "Bulldog ",
    "lying ",
    "on ",
    "its ",
    "side ",
    "on ",
    "a ",
    "blue ",
    "wooden ",
    "",
    "floor. ",
    "The ",
    "",
    "",
    "dog's ",
    "head ",
    "is ",
    "resting ",
    "on ",
    "its ",
    "",
    "paws, ",
    "and ",
    "it ",
    "appears ",
    "to ",
    "be ",
    "",
    "sleeping. ",
    "The ",
    "meme ",
    "has ",
    "a ",
    "text ",
    "overlay ",
    "that ",
    "reads ",
    "",
    "",
    "",
    "\"MONDAY. ",
    "",
    "",
    "",
    "",
    "JUST...MONDAY.\" ",
    "This ",
    "suggests ",
    "that ",
    "the ",
    "meme ",
    "is ",
    "making ",
    "a ",
    "humorous ",
    "comment ",
    "about ",
    "the ",
    "start ",
    "of ",
    "the ",
    "work ",
    "",
    "week. ",
    "The ",
    "",
    "",
    "dog's ",
    "relaxed ",
    "posture ",
    "and ",
    "the ",
    "text ",
    "overlay ",
    "together ",
    "create ",
    "a ",
    "",
    "",
    "light-hearted ",
    "and ",
    "",
    "relatable ",
    "",
    "",
    "message."
  ],
  "started_at": "2024-03-11T19:58:46.441993Z",
  "status": "succeeded",
  "urls": {
    "stream": "https://streaming-api.svc.us.c.replicate.net/v1/streams/vwnyhw22itkrswhzyd3irptyibxlukuw4ycgoojpm42dw3wfinsq",
    "get": "https://api.replicate.com/v1/predictions/xfwnzedbxhtwqftqabcj5aiaf4",
    "cancel": "https://api.replicate.com/v1/predictions/xfwnzedbxhtwqftqabcj5aiaf4/cancel"
  },
  "version": "d1823e6f68cd3d57f2d315b9357dfa85f53817120ae0de8d2b95fbc8e93a1385"
}

Generated in

5.7 seconds

Tweak it Share Report