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

Internlm

InternLM is a series of large language models accompanied by a public code repository, model release links, model cards, technical information, a chat web interface and API documentation.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Internlm — 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 the repository and linked resources when selecting an InternLM release, reviewing its model information, or identifying a chat or API route for an application.

Who it is for

  • Developers evaluating language models for conversational applications.
  • Researchers and practitioners exploring model releases, fine-tuning or long-context text tasks.

Documented uses

  • Compare InternLM model releases and find their associated model resources.
  • Try conversational use through the linked chat interface.
  • Explore API integration or self-hosted use of an InternLM model.
Before you choose

Requirements & limits

The repository is public and identifies the project license as Apache-2.0. It links to a chat web interface and API documentation.

  • Releases vary by model size and tuning, so consult the relevant model card when selecting a checkpoint.
  • InternLM3-8B-Instruct is described as supporting both normal response and deep-thinking modes.
  • The repository is Python-based and links model resources through Hugging Face and other platforms.

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

Run InternLM3-8B-Instruct conversation inference with Transformers

Load InternLM3-8B-Instruct and generate a response to a conversational prompt locally.

Before you start

  • Python >= 3.8
  • PyTorch >= 1.12.0 (2.0.0 and above recommended)
  • Transformers >= 4.38
  • A device with enough memory to load the model; the README warns that float32 loading might cause an out-of-memory error and notes that 4-bit loading uses nearly 8 GB of GPU memory.

Steps

  1. Load the tokenizer for internlm/internlm3-8b-instruct with trust_remote_code=True.

  2. Load the model with AutoModelForCausalLM.from_pretrained, trust_remote_code=True, and torch_dtype=torch.float16, then set it to evaluation mode.

  3. Prepare system and user messages, and tokenize them using tokenizer.apply_chat_template with add_generation_prompt=True and return_tensors="pt".

  4. Generate with model.generate, remove the input token IDs from the generated IDs, and decode the response with tokenizer.batch_decode.

Expected outcome

The example decodes and prints the prompt and generated response.

Guide limitations

  • The README cautions that loading as float32 might cause an out-of-memory error.
  • For low-resource devices, the README describes optional 4-bit or 8-bit loading with bitsandbytes; it estimates nearly 8 GB of GPU memory for 4-bit InternLM3 8B.
  • The example documents local Transformers inference, not a hosted deployment.
Documentation 1 ↗
Source-based practical guide

Run local batch inference with InternLM3-8B-Instruct using LMDeploy

Generate a response locally by passing a prompt to an LMDeploy pipeline.

Before you start

  • The documented setup installs LMDeploy with pip install lmdeploy.
  • The example uses the model internlm/internlm3-8b-instruct.

Steps

  1. Install LMDeploy using the command below.

    Documented command
    pip install lmdeploy

  2. Create an LMDeploy pipeline for internlm/internlm3-8b-instruct.

  3. Pass a prompt to the pipeline and print the returned response.

Expected outcome

The example prints a response to the prompt.

Guide limitations

  • The recipe documents local batch inference only; it does not specify hardware requirements or provide a performance guarantee.
Documentation 1 ↗
Evidence and transparency

Sources & review

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