OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Oceangpt

Oceangpt is an open-source project that develops large language models for ocean-science tasks and provides model checkpoints, documentation and example inference code.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Oceangpt — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use Oceangpt when you want to evaluate or build with models targeted at ocean-science tasks, especially marine question answering, image interpretation or MOOS code generation.

Who it is for

  • Marine and ocean-science researchers exploring language-model assistance for domain questions.
  • Developers building marine question-answering or ocean-science image-interpretation workflows.
  • Robotics developers investigating MOOS code generation with the OceanGPT-coder model.

Documented uses

  • Ask questions about marine expertise using an OceanGPT basic model.
  • Interpret marine-science or sonar images with OceanGPT-o.
  • Generate MOOS code for underwater-robot workflows with OceanGPT-coder.
  • Download a checkpoint and run inference with the documented Python examples or supported inference tools.
  • Fine-tune a model for a task-oriented question-answering application using the linked guide.
Before you choose

Requirements & limits

The repository is public and identifies its license as MIT. The README documents local setup and links to model downloads on Hugging Face and ModelScope.

  • Choose the checkpoint by task: basic variants are presented for marine Q&A, OceanGPT-o for image interpretation, and OceanGPT-coder for MOOS code generation.
  • The setup path uses Python 3.11, Conda and repository requirements; local inference also depends on the selected model and available hardware.
  • The project provides a separate Gradio interface and separate inference-framework examples; these are tools for using the models, not the model checkpoints themselves.
  • The repository notes that updates can change model capabilities, so check the selected checkpoint's current documentation.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up the documented Python environment and download a model

Prepare the repository environment and obtain an OceanGPT model checkpoint for local inference.

Before you start

  • The OceanGPT repository
  • Conda
  • Python 3.11
  • Git LFS or the Hugging Face Hub or ModelScope download tooling

Steps

  1. Create and activate a Conda environment with Python 3.11.

  2. Install the repository dependencies with the command below.

    Documented command
    pip install -r requirements.txt

  3. Download a model from Hugging Face using Git LFS or the Hugging Face CLI, or use the documented ModelScope options.

Expected outcome

A Python environment with the selected model downloaded, ready for the repository's model-specific inference instructions.

Guide limitations

  • This guide covers environment setup and model download only; hardware requirements and successful inference are not specified here.
  • Model capabilities vary by checkpoint and may change between releases.
Documentation 1 ↗
Source-based practical guide

Configure and open the local Gradio interface

Prepare the provided local web interface to interact with a model already stored on your machine.

Before you start

  • A local OceanGPT, OceanGPT-o or OceanGPT-coder model
  • The repository's app.py and its installed requirements

Steps

  1. Set the model path in app.py to the local model directory.

  2. Run the command below from the repository.

    Documented command
    python app.py

  3. Open the local interface in a browser at the address and port specified in the README.

Expected outcome

The repository's interactive local interface is launched for model interaction.

Guide limitations

  • A local model is required, and the README does not specify hardware requirements.
Documentation 1 ↗
COMPARE OPTIONS

Compare options for Model inference and serving

These projects document a common task and share a project role. This is a source-based comparison, not an execution benchmark or a guarantee of interchangeability.

ProjectDocumented purposePlatformAccess and requirements
OceangptAsk questions about marine expertise using an OceanGPT basic model.The repository is public and identifies its license as MIT. The README documents local setup and links to model downloads on Hugging Face and ModelScope.
Pulse Of MotionEstimate a video's physical frame rate from its visual dynamics.LinuxThe repository is public. Its README describes a released checkpoint that downloads from Hugging Face on first inference, with a manual download option. The strongest internal model is not fully released; the README directs interested users to contact the project about access. No license information is provided in the repository evidence.
Neural Image AssessmentTrain the model using AVA images and rating annotations.Cloud, Linux, macOS, Windows
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
12 October 2026
Original website title
OceanGPT/OceanGPT · Official homepage title · retained in its source language
How we review sourcesSuggest a correction