DeepSeek VL 7B on Replicate API

DeepSeek VL · DeepSeek

ServerlessOpen Weights

Last refreshed 2026-09-18. Next refresh: weekly.

Why use DeepSeek VL 7B on Replicate API?

Replicate API offers DeepSeek VL 7B with pay-as-you-go pricing at $0.05/1M input tokens. Replicate is a cloud-based platform that enables users to run machine learning models easily and efficiently.

Input / 1M
$0.050
Output / 1M
$0.25
Cache
Not sourced
Batch
Not sourced

Setup recipe

Python + curl
Install
pip install replicate
Auth
export REPLICATE_API_TOKEN=...
Call
import replicate
output = replicate.run(
    "deepseek-vl-7b",
    input={"prompt": "Hello"}
Model ID
deepseek-vl-7b

Request example

import replicate

# reads REPLICATE_API_TOKEN from env
# deepseek-vl-7b format: "owner/model-name" (latest version) or "owner/model-name:version-hash"
output = replicate.run(
    "deepseek-vl-7b",
    input={"prompt": "Hello"}
)
# Output is a list or generator depending on the model
print("".join(output))

Gotchas

  • Replicate uses "owner/model-name" format (e.g. "meta/meta-llama-3-8b-instruct") for the latest version, or "owner/model-name:version-sha" to pin to a specific version. The REST endpoint splits owner and model-name into the path: /v1/models/{owner}/{model-name}/predictions.
  • The examples expect REPLICATE_API_TOKEN; rename it only if your application config maps the new variable.

Pricing

TypePrice (per 1M)
Input tokens$0.05
Output tokens$0.25

Capabilities

VisionMultimodal

About DeepSeek VL 7B

DeepSeek-VL 7B is an open-source vision-language model engineered for robust real-world applications. With its general multimodal understanding capabilities, it can process logical diagrams, web pages, formulas, scientific literature, natural images, and complex scenarios involving embodied intelligence. The model features a hybrid vision encoder that integrates SigLIP-L and SAM-B, allowing it to handle high-resolution (1024 x 1024) image inputs. Built on the DeepSeek-LLM-7b-base foundation, it is pre-trained on roughly 2 trillion text tokens and further trained on approximately 400 billion vision-language tokens.

Get Started

Model Specs

Released2024-03-15
Parameters7B
ArchitectureDecoder Only