Project AURIONAI Use-Cases Research & Innovation
reasoning

DeepSeekMath 7B Instruct

7B model further pre-trained on 120 billion maths tokens and tuned for step-by-step mathematical reasoning; introduced the GRPO training method.

DeepSeekDeepSeekChina
DeepSeek
Levelmodel
Classreasoning
Parametersnot in the notes
Contextnot in the notes
Released5 Feb 2024
Opennessopen weights restricted
Runs onnot in the notes
Statusactive

R&D Topics

Mathematical reasoning and formal theorem provingexemplar

Identity

Developer
DeepSeek
Family
not in the notes
Released
5 Feb 2024 sourcesregistry date
Weights published
–
Record checked
12 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.

Architecture and Size

Architecture
dense transformer
Total parameters
– sources
Active parameters
–
Checkpoint parameters
–
Context length
not in the notes
Input
text
Output
text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).

Openness and Access

Openness
open weights restricted
Licence
DeepSeek License Agreement 1.0 sources
Access routes
open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.

Deployment

Runs on
not in the notes
Footprint
not in the notes
Runtimes
not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.

Indonesia

Indonesian support
not stated
Regional languages
not in the notes
Jakarta hosting
unknown
Languages
not in the notes
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.

Benchmark Results

The notes record no benchmark result for this model.

Lineage

The notes record no lineage for this model.

Industry Use Cases

No industry use case links this model.

Sources

  1. github.com/deepseek-ai/DeepSeek-Math https://github.com/deepseek-ai/DeepSeek-Math
  2. huggingface.co/deepseek-ai/deepseek-math-7b-instruct https://huggingface.co/deepseek-ai/deepseek-math-7b-instruct
  3. kaggle.com/models/deepseek-ai/deepseek-math https://www.kaggle.com/models/deepseek-ai/deepseek-math
  4. kaggle.com/models/deepseek-ai/deepseek-prover https://www.kaggle.com/models/deepseek-ai/deepseek-prover

To cite this page: Project AURION, AI Use-Cases Research & Innovation, "DeepSeekMath 7B Instruct", https://aurion.semai.id/models/deepseek-deepseek-math-7b-instruct/, records of October 2026.