History-Llms

Visit Tool

history-llms is a Research & Education tool that trains large language models on historical data. It provides time-locked models that act as windows into the past for research in humanities, social sciences, and computer science.

Claim this tool

2Views

At a glance

Pricing

Open Source

Free tier

Yes

API

Skill level

Technical

About

What is history-llms?

history-llms is an information hub for a project focused on training the largest possible historical Large Language Models (LLMs). These models are designed to be fully time-locked, meaning they only access information up to a specific knowledge-cutoff date, such as 1913, 1929, or 1946. This approach allows researchers to explore historical discourse patterns without the hindsight contamination present in modern LLMs. The project aims to provide tools for exploring massive textual corpora and complements traditional archival research, serving as a window into past perspectives on various topics. The models are intended for scientific applications, enabling research in humanities, social sciences, and computer science, with a commitment to minimizing interference with the normative judgments acquired during pretraining.

Best used for

Ideal for professors and researchers who need to analyze historical texts, understand past societal views, and explore the boundaries of thought in specific historical periods. Especially valuable for studying how historical events and ideas were perceived before their outcomes were known.

Common actions

analyze historical data

research historical perspectives

explore past discourse

workflowsopen-sourcedeepfakecollaborationgithub copilotautomated workflowface swappinglow-code/no-code"AI Agents"

Capabilities

Key features

Time-locked historical models
Qwen3 architecture
80B tokens historical data
Specific knowledge cutoffs
Uncontaminated bootstrapping

Target Audience

professorresearcher

Integrations

Not yet documented

Pricing & Plans

Open Source

Free

FAQs

What are 'time-locked' historical LLMs?

Time-locked historical LLMs are models trained exclusively on texts published before specific cutoff dates (e.g., 1913, 1929). This ensures they do not have access to any information beyond their knowledge-cutoff date, providing an authentic window into the textual culture and perspectives of that era without hindsight contamination.

Why not just use modern LLMs like GPT-5 for historical roleplaying?

Modern LLMs suffer from hindsight contamination; they know how history unfolds. Even when instructed to roleplay, their inherent knowledge shapes responses. Time-locked models genuinely embody their training data, offering insights into what was thinkable or predictable in a given moment, which modern LLMs cannot truly replicate.

Will these models reproduce historical biases and sensitive content?

Yes, historical texts contain racism, antisemitism, misogyny, and imperialist views, and the models will reproduce these. This is considered a crucial feature, as understanding how such views were articulated and normalized in the past is essential for historical research and analysis.

Trending

Subcategories trending in Research & Education

Study Assistants Knowledge Management Course Creation Scientific Computing Summarization Language Learning

Trending

Explore

Browse AI tools by category

Content & Design Productivity & Business Coding & Development AI Agents & Automation Research & Education Wellness & Lifestyle Career Development Marketing & Growth Data & Analytics Customer Support & CX Finance E-commerce