ashoknimiwal/DeepSeek-R1-14B-Research-Snapshot

TEXT GENERATIONConcurrency Cost:1Model Size:14.8BQuant:FP8Ctx Length:32kPublished:Apr 29, 2026License:mitArchitecture:Transformer Open Weights Cold

The ashoknimiwal/DeepSeek-R1-14B-Research-Snapshot is an unmodified 14.8 billion parameter language model based on the DeepSeek-R1-Distill-Qwen architecture, with a 32768 token context length. This specific snapshot, created by ashoknimiwal, is preserved for research reproducibility, ensuring an identical model artifact for future studies. It is primarily intended for research purposes, particularly in contexts requiring a fixed model version for comparative analysis or replication of results. The model is tagged for research reproducibility, corporate culture, text classification, NLP, earnings calls, and reasoning tasks.

Loading preview...

DeepSeek-R1-14B-Research-Snapshot Overview

This model, ashoknimiwal/DeepSeek-R1-14B-Research-Snapshot, is an exact, unmodified copy of the deepseek-ai/DeepSeek-R1-Distill-Qwen-14B model. It features 14.8 billion parameters and supports a context length of 32768 tokens. The primary purpose of this snapshot is to ensure research reproducibility, providing a stable and identical model artifact for researchers.

Key Characteristics

  • Unmodified Snapshot: This repository contains the model as it was downloaded on April 29, 2026, with no alterations to its weights, tokenizer files, or configurations.
  • Research Focus: Explicitly created to support an associated manuscript (currently under review), ensuring that the model used in research can be precisely replicated.
  • Base Architecture: Built upon the DeepSeek-R1-Distill-Qwen architecture, indicating its foundation in advanced language modeling techniques.
  • Tags: Relevant tags include research-reproducibility, corporate-culture, text-classification, nlp, earnings-calls, deepseek, qwen2.5, and reasoning.

Intended Use Cases

This model is particularly well-suited for:

  • Reproducing Research: Ideal for researchers needing to verify or build upon studies that utilized the original deepseek-ai/DeepSeek-R1-Distill-Qwen-14B model at a specific point in time.
  • Comparative Analysis: Useful for benchmarking against other models where a fixed, known baseline is required.
  • Academic Studies: Supports academic work where model versioning and immutability are critical for methodological rigor.

Methodological details and results related to its specific research application will be made available upon the acceptance of the associated manuscript.