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

Chinese Llava Med

Chinese Llava Med is a Chinese-language medical vision-and-language model project whose repository provides image-analysis examples, training and evaluation datasets, fine-tuning configurations, evaluation scripts and a LLaMA-Factory inference setup.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Chinese Llava Med — 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 it when you need a repository-based workflow for experimenting with Chinese medical image question answering, rather than a hosted clinical application.

Who it is for

  • Researchers and developers building or fine-tuning Chinese medical vision-language models.
  • Teams evaluating model responses on medical image question-answering tasks.

Documented uses

  • Explore model responses to tissue-slide, chest X-ray and CT images.
  • Fine-tune a medical multimodal model using the supplied instruction dataset.
  • Generate and score responses against the supplied evaluation set.
  • Run an interactive inference chat through LLaMA-Factory.
Before you choose

Requirements & limits

The repository is public and lists an Apache-2.0 license.

  • The README's full fine-tuning example uses distributed PyTorch with two CUDA devices.
  • The supplied datasets are a 60,000-item instruction-tuning set and a 115-item evaluation set.
  • The reported benchmark score is tied to the named evaluation set and scoring approach.

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 LLaMA-Factory inference chat

Open an interactive chat interface using the final model and the repository's inference configuration.

Before you start

  • The Chinese Llava Med repository and its inference configuration.
  • LLaMA-Factory installed in the environment.

Steps

  1. Install LLaMA-Factory using the README's documented installation command.

  2. Run llamafactory-cli webchat config/llava1_5_full_sft_infer.yaml from the repository context.

Expected outcome

The documented command starts an interactive webchat interface using the inference configuration.

Guide limitations

  • The README describes the interface but not deployment beyond it.
Documentation 1 ↗
Source-based practical guide

Fine-tune and export a medical vision-language model

Follow the repository's documented full fine-tuning and export stages with LLaMA-Factory.

Before you start

  • LLaMA-Factory installed with the documented torch and metrics extras.
  • The repository's config/llava1_5_full_sft.yaml and config/llava1_5_full_sft_export.yaml files.
  • Two CUDA devices for the README's full fine-tuning example.

Steps

  1. Install LLaMA-Factory using the README's documented commands.

  2. Run the distributed training command with config/llava1_5_full_sft.yaml.

  3. Set export_hub_model_id and hf_hub_token in config/llava1_5_full_sft_export.yaml.

  4. Run the documented LLaMA-Factory export command.

Expected outcome

The documented workflow trains a model and proceeds to its export stage.

Guide limitations

  • The README also offers a LoRA YAML configuration, but its detailed execution steps are not shown in the supplied instructions.
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
Chinese Llava MedExplore model responses to tissue-slide, chest X-ray and CT images.CLI, LinuxThe repository is public and lists an Apache-2.0 license.
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
16 October 2026
Original website title
BUAADreamer/Chinese-LLaVA-Med · Official homepage title · retained in its source language
How we review sourcesSuggest a correction