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

Dots.llm1

Dots.llm1 is a family of base and instruction-tuned large language models, with checkpoints hosted through Hugging Face and documented options for text generation and model serving.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Dots.llm1 — 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 Dots.llm1 when you want to explore its base and instruct variants, generate text with a documented Transformers workflow, or set up a serving workflow using vLLM or SGLang.

Who it is for

  • Developers building text-generation or chat prototypes from language-model checkpoints.
  • Researchers comparing base and instruction-tuned models or studying intermediate training checkpoints.

Documented uses

  • Generate a continuation from a text prompt with the base model.
  • Produce chat-style responses with the instruct model.
  • Serve the instruct checkpoint through vLLM or SGLang.
  • Explore intermediate checkpoints for research into model training dynamics.
Before you choose

Requirements & limits

The repository lists an MIT license and links to Hugging Face checkpoints and a demo.

  • The repository lists 142 billion total parameters and 14 billion activated parameters.
  • The stated context length is 32,768 tokens, with English and Chinese listed as supported languages.
  • Documented inference routes include Transformers 4.53.0 or later, vLLM 0.9.2 or later, and SGLang.
  • The README serving examples use tensor parallelism of eight; it does not provide hardware requirements.

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

Generate text with a dots.llm1 checkpoint using Transformers

Load a base or instruction-tuned checkpoint and generate text using the README's Transformers examples.

Before you start

  • Transformers version 4.53.0 or later.
  • PyTorch, as used in the examples.
  • Access to the relevant Hugging Face checkpoint.

Steps

  1. Choose the base checkpoint for text completion or the instruct checkpoint for chat completion.

  2. Load the checkpoint tokenizer and model with Transformers; the examples use bfloat16 and automatic device mapping.

  3. Prepare a prompt or chat message, tokenize it, and call model.generate.

  4. Decode the generated tokens to obtain the text output.

Expected outcome

Text generated from the selected checkpoint, either as a continuation or a chat-style response.

Guide limitations

  • The README examples do not specify hardware requirements.
  • The Transformers examples use bfloat16; the repository does not describe alternative precision settings.
Documentation 1 ↗
Source-based practical guide

Serve the instruct checkpoint with vLLM

Launch the instruct checkpoint through vLLM and expose the documented OpenAI-compatible API endpoint.

Before you start

  • vLLM version 0.9.2 or later.
  • The dots.llm1 instruct checkpoint.
  • A serving environment configured for tensor parallelism of eight, as in the README example.

Steps

  1. Start vLLM with the instruct checkpoint, port 8000, and tensor parallelism set to eight.

  2. Send requests to the documented OpenAI-compatible API at http://localhost:8000/v1.

Expected outcome

A vLLM server for the instruct checkpoint with an OpenAI-compatible API endpoint.

Guide limitations

  • The README does not specify the hardware needed for the example configuration.
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
Dots.llm1Generate a continuation from a text prompt with the base model.CloudThe repository lists an MIT license and links to Hugging Face checkpoints and a demo.
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
14 October 2026
Original website title
studio-dots-ai/dots.llm1 · Official homepage title · retained in its source language
How we review sourcesSuggest a correction