Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
February 22, 2025
·
Seattle
Wanderheart
Former Wizards of the Coast developers explain building an AI Dungeon Master for D&D, addressing core LLM challenges to deliver reliable, engaging gameplay.
Video
Overview
“Dungeons & Dragons” with an AI Dungeon Master, built by former Wizards of the Coast game developers.
Links
WanderHeart: AI Game Master for dynamic, no-prep, multiplayer TTRPGs.
Tech stack
- GPT-4GPT-4 is OpenAI’s large multimodal model: it processes both text and image inputs, delivering human-level performance on complex professional and academic benchmarks.This is OpenAI’s latest milestone in scaling deep learning: a large multimodal model accepting both text and image inputs. It demonstrates a significant capability leap over its predecessor, scoring in the top 10% on a simulated bar exam (GPT-3.5 scored in the bottom 10%). The model handles nuanced instructions and long-form content, supporting context windows up to 32,768 tokens (32K model). This capacity allows processing up to 25,000 words in a single, complex prompt. GPT-4 is engineered for enhanced reliability, steerability, and advanced reasoning across diverse tasks.
- AI Dungeon MasterThe AI Dungeon Master: a cutting-edge LLM application that generates, narrates, and manages dynamic, rule-compliant campaigns (e.g., D&D 5e) in real-time.This technology serves as a virtual game facilitator, leveraging advanced AI models to deliver a complete tabletop RPG experience. It handles all core DM functions: dynamic story generation, rule adjudication (dice rolls, combat mechanics), and world consistency. The system adapts instantly to player input, weaving custom lore and character actions into an infinite narrative. Specific implementations (like Fables.gg) integrate features such as world-building tools and tactical virtual tabletops, ensuring a balanced, engaging, and highly personalized adventure for solo or group play.
- GPT-3A 175-billion parameter autoregressive language model that masters complex tasks through few-shot learning.OpenAI debuted GPT-3 in 2020: a transformer-based engine trained on 570GB of filtered text. It utilizes 175 billion parameters to execute diverse functions (including Python scripting and logical reasoning) using only natural language prompts. This architecture removed the requirement for task-specific fine-tuning: establishing the foundation for modern tools like GitHub Copilot and the initial ChatGPT release.
- Llama-2Llama 2 is Meta AI's powerful, openly accessible family of large language models (LLMs), featuring models from 7B to 70B parameters for research and commercial applications.Llama 2 is Meta AI's next-generation LLM family, released for free research and commercial use. The collection includes both pre-trained foundation models and instruction-tuned 'Chat' variants, scaling from 7 billion (7B) up to 70 billion (70B) parameters. Key technical upgrades over Llama 1 involve training on 2 trillion tokens (40% more data) and doubling the context length to 4096 tokens. The Llama-2-chat models were rigorously aligned using Reinforcement Learning from Human Feedback (RLHF), positioning them as a top-tier, openly available option for developers building advanced generative AI solutions.
- PaLM 2Google's versatile large language model optimized for advanced reasoning, multilingual translation, and coding across four distinct scales.PaLM 2 powers 25+ Google products (including Gemini and Workspace) using a Transformer-based architecture trained on a massive corpus of 100+ languages. It excels in specialized tasks: solving complex math problems, generating high-quality code, and passing professional-level exams. Developers deploy the model via the PaLM API in four sizes: Gecko, Otter, Bison, and Unicorn. Gecko is lightweight enough to run locally on mobile devices (offline), while Unicorn handles the most complex, data-heavy reasoning tasks at scale.
- BLOOMA 176-billion parameter open-access multilingual language model built by the BigScience research collective.BLOOM is the result of a year-long collaboration involving 1,000+ researchers from 70+ countries. It supports 46 natural languages and 13 programming languages: it provides a high-performance alternative to proprietary models. The model was trained on the Jean Zay supercomputer in France using the 1.6-terabyte ROOTS dataset (a massive collection of diverse text sources). By providing full access to its weights and training process, BLOOM enables global developers to build and audit AI tools without the restrictions of closed-door APIs.
- BERTBERT (Bidirectional Encoder Representations from Transformers) is a foundational, pre-trained NLP model that uses a Transformer encoder to process text bidirectionally, capturing full word context for superior language understanding.BERT is a revolutionary language representation model introduced by Google AI Language in 2018. It is built on the Transformer architecture and distinguishes itself by being deeply bidirectional: it processes the entire sequence of words (left and right context) simultaneously, unlike previous unidirectional models. This capability is achieved through a Masked Language Model (MLM) pre-training objective. The model, released in sizes like BERTBASE (110 million parameters) and BERTLARGE (340 million parameters), dramatically improved the state-of-the-art across 11+ Natural Language Processing tasks, including question answering (SQuAD) and sentiment analysis, establishing a new baseline for the field.
- RoBERTaRoBERTa (Robustly Optimized BERT Pretraining Approach) is a high-performance language model from Facebook AI that significantly outperforms BERT by optimizing the pretraining strategy, not the core architecture.RoBERTa is a robustly optimized version of the BERT model, developed by researchers at Facebook AI in 2019. The team conducted a replication study, proving BERT was undertrained and could achieve state-of-the-art results with a refined recipe: they removed the Next Sentence Prediction (NSP) objective, implemented dynamic masking, and scaled up training dramatically. Specifically, RoBERTa trained for 500K steps (up from 100K) on a massive 160GB of text data (ten times BERT’s data) using much larger batch sizes (up to 8K). This optimized approach yielded superior performance on major benchmarks like GLUE, RACE, and SQuAD, establishing RoBERTa as a benchmark for subsequent language model development.
Related projects
AI Adventure
Seattle
Demo shows how to build an open‑source AI role‑playing game that generates story, images, and audio on the…
Harnessing LLM's For Hard Gameplay Dynamics
Seattle
This talk demonstrates using RAG and game logic to create a detective game where players expose NPC lies…
Using LLMs for real-world robot control
Seattle
This talk demonstrates controlling a ground robot using large language models with OpenAI’s function calling, including a live…
Leela Quest
Seattle
Explore how an AI‑enhanced version of the ancient Leela board game guides participants through structured self‑discovery, blending spirituality…
Bootstrapping AI To Success
Seattle
Learn how we built the AI Game Master from concept to revenue, detailing model choices, infrastructure decisions, and…
Momentic
Seattle
The talk explores deploying AI agents to test web apps by interpreting user flows in natural language and…