# NanoGPT Speedrun **Entity class:** Benchmark and software ecosystem **Domain:** Artificial intelligence / Model training **Maturity:** Developed ## Definition The NanoGPT speedrun is a collaborative effort to minimize the time required to train a GPT-2-scale model to a fixed performance target on specified hardware and data. ## Mechanism and significance Because the target remains fixed, changes in training time expose algorithmic, architectural, optimizer, and systems gains that are otherwise obscured by scaling model size. ## Relationships - **Research dossier:** [[research/ASI and RSI Timeline Research Moonshots|ASI and RSI Timeline Research Moonshots]] - **Ontology route:** [[ASI and RSI Timeline Ontology#Model Architecture and Compute Economics|Model Architecture and Compute Economics]] - **Primary fields:** [[Machine Intelligence]] · [[AI Infrastructure]] · [[Compute Race]] - **Adjacent concepts:** [[Model Self-Redesign]] · [[nanochat]] · [[Algorithmic Improvement Without Data Scaling]] · [[Compact Reasoning Kernel]] - **People and software:** [[Keller Jordan]] · [[Andrej Karpathy]] · [[NanoGPT]] ## Sources and provenance - [[research/ASI and RSI Timeline Research Moonshots|ASI and RSI Timeline Research Moonshots]] — immediate source for this node's role in the broadcast research map. - [modded-nanogpt repository](https://github.com/KellerJordan/modded-nanogpt) — primary organizational or project source. ## Evidence boundary The research dossier establishes why this entity or concept belongs in the Moonshots ontology. Time-sensitive organizational, product, policy, and performance claims should be checked against the linked primary source or a current authoritative source before reuse as settled fact.