Embeds multi-head self-attention inside an LSTM-style memory, so stored memories can attend to one another instead of just sitting in separate slots — sharpening relational reasoning and topping WikiText-103, Project Gutenberg, and GigaWord.
Introduced the two-stage recipe behind the GPT lineage: unsupervised generative pre-training on unlabeled text, then supervised fine-tuning per task. A single 12-layer Transformer decoder beat bespoke architectures on 9 of 12 NLP benchmarks.
AI Deploy2018
Converts, quantizes, and runs deep learning models from PyTorch, TensorFlow, ONNX, and PaddlePaddle across Intel CPUs, GPUs, and NPUs without the training framework. Adds a GenAI pipeline for LLMs plus Hugging Face, vLLM, and LangChain integrations.
AI Model2018
Provides unified model definitions and a single API for pretrained text, vision, audio, and multimodal models for both training and inference. Emphasizes cross-framework compatibility (PyTorch/TF/JAX), pipeline-based inference, and direct access to 1M+ Hub checkpoints.
AI Model2018
Turns model definitions into a shared layer across training and inference stacks, covering text, vision, audio, video, and multimodal models. Pipelines, Trainer, and generation APIs make pretrained models usable without locking teams to one framework.
AI Deploy2019
Turns Python ML code into production inference APIs that scale on Kubernetes or any cloud. Bundles models, dependencies, and serving logic into versioned "Bentos" with autoscaling, scale-to-zero, and multi-GPU serving for LLMs and custom models.
A 1.5B-parameter model trained only to predict the next token on diverse web text does translation, summarization, and QA zero-shot, with no fine-tuning. It recast NLP tasks as conditional language modeling and sparked the staged-release misuse debate.
AI Agent2019
Open-source Airtable alternative for building databases, apps, automations, and AI agents without code over a PostgreSQL-backed REST API. The Kuma assistant turns plain language into tables and workflows; self-hostable with full data ownership.
AI Train2019
Train and experiment with multi-billion to trillion-parameter transformer models on large GPU clusters using GPU-optimized building blocks and reference training scripts; offers advanced parallelism and mixed-precision support for research teams and ML engineers.
Demonstrated that language model loss falls as a smooth power law in model size, data, and compute across more than seven orders of magnitude — turning "make it bigger" from a hunch into a budget you can plan, and justifying the GPT-3 scale-up.
At 175 billion parameters, this autoregressive model becomes a strong few-shot learner: it handles translation, QA, and reasoning from a few prompt examples with no gradient updates, establishing in-context learning as an alternative to fine-tuning.