The publication AI reads — a remote MCP server, nothing to install.Connect your agent →
Model register·internlm

internlm/Atria-Dawn-Preview

internlm/Atria-Dawn-Preview is a internlm record on this register. The safetensors index gives an exact parameter count of 753,329,940,480. Architectural context length (max_position_embeddings) is 1,048,576. The listed licence is mit. The Hub repository was created on 11 Sept 2026, which is the release date this register uses.

Primary source: the Hugging Face model card for internlm/Atria-Dawn-Preview. Fields below come from that card, the Hub model API, and the repository’s own config.json when those files exist. A field we cannot source is omitted rather than guessed.

Organisationinternlm
ReleasedRepo creation date on Hugging Face11 Sept 2026
Last modified16 Sept 2026
ParametersExact, from the safetensors index753.3B 753,329,940,480
Context lengthmax_position_embeddings — architectural maximum, not necessarily served1M 1,048,576
Licencemit
File formatssafetensors
Architectureglm_moe_dsa
Layers78
Vocabulary154,880
DownloadsAt last check1.2k
Likes236
GatedNo

How to run

Safetensors weights are on the Hugging Face repository. The parameter count on this page came from that index.

Load it from huggingface.co/internlm/Atria-Dawn-Preview. This page does not add install steps that the Hub files do not state.

Agent capabilities

Derived from the model’s own chat template, with the match counts behind each finding.

  • ●Tool calling16 matches in template
  • ●Reasoning / thinking17 matches in template
  • ●System prompt1 match in template
  • ●Image input2 matches in template
  • ●Video input1 match in template
  • ●Audio input1 match in template

Questions this record can answer

When was Atria-Dawn-Preview released?
The Hugging Face repository was created on 11 Sept 2026, which is the date this register uses.
How many parameters does Atria-Dawn-Preview have?
The safetensors index gives 753,329,940,480 parameters, shown here as 753.3B. That figure is exact, not rounded.
What is the context length of Atria-Dawn-Preview?
1,048,576 tokens. That is max_position_embeddings from the model's own config.json — the architectural maximum, not necessarily the context a provider serves.
What licence is Atria-Dawn-Preview under?
The Hub card lists mit.
Is Atria-Dawn-Preview gated?
No. The Hugging Face model API does not mark this repository as gated.
Does the Atria-Dawn-Preview chat template show tool calling?
The chat template contains evidence of tool calling.

Where each field came from

  • Hugging Face model API — releasedAt, lastModified, downloads, likes, task, license, gated, parameters. Checked 26 Sept 2026.
  • config.json — contextLength, layers, vocabSize, architecture. Checked 26 Sept 2026.
  • chat template — capabilities. Checked 26 Sept 2026.

First recorded 24 Sept 2026, last checked 26 Sept 2026. This record is compiled automatically from the sources above. Where a field is blank, we could not source it and have not guessed. Canonical repository: huggingface.co/internlm/Atria-Dawn-Preview.