The AI terms that
actually matter.
Clear, technical definitions of 113 key concepts — written by AI engineers, not marketers.
dGSLM (Dialogue Generative Spoken Language Model)
dGSLM is a full-duplex generative spoken language model that uses a Siamese dual-encoder architecture with cross-attention to model both sides of a spoken dialogue simultaneously, without any text in the loop.
Read definitionDiffusionGemma
An experimental, open weights multimodal model by Google DeepMind that utilizes discrete text diffusion instead of autoregression to generate 256 token blocks in parallel. This enables extreme inference speeds.
Read definitionE-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness)
E-E-A-T is the four-factor framework in Google's Search Quality Rater Guidelines that human raters use to judge content and creator quality, with Trust as the load-bearing factor, and it has become a proxy signal that AI Overviews and other generative search engines lean on when deciding which sources to cite.
Read definitionEmbeddings
Embeddings are dense numerical vectors that represent the meaning of text, images, or other data in a high-dimensional space — where items that are semantically similar end up geometrically close together, enabling AI systems to measure meaning rather than just match keywords.
Read definitionFlash Attention
Flash Attention is an I/O-aware, exact attention algorithm that fundamentally solves the memory wall in Transformer models. By fusing operations and minimizing costly reads/writes between the GPU's High Bandwidth Memory (HBM) and SRAM, it significantly speeds up processing and enables massive context windows.
Read definitionGemma 3n E4B
Gemma 3n E4B is Google's mobile-first, multimodal (text, image, audio) model that uses a nested MatFormer architecture and Per-Layer Embeddings to run an 8-billion-parameter model in roughly 3GB of accelerator memory, making it one of the cheapest ways to get capable multimodal inference, whether served via a hosted API at a fraction of median pricing or run entirely on-device for zero marginal cost.
Read definitionGemma 4
Gemma 4 is Google DeepMind's fourth-generation family of open-weight multimodal AI models, spanning four sizes (E2B, E4B, 26B MoE, 31B Dense), built from the same research as Gemini 3 and designed for advanced reasoning, agentic workflows, and on-device deployment.
Read definitionGenerative AI (Gen AI)
Generative AI is a branch of artificial intelligence that creates new content or data patterns mimicking human-like creativity, based on learned information from existing datasets.
Read definitionGenerative Engine Optimization (GEO)
Generative Engine Optimization (GEO) is the process of optimizing content to be discovered, retrieved, and cited by AI-driven generative search engines like Perplexity, ChatGPT, and Google AI Overviews.
Read definitionGLM-5.1
GLM-5.1 is an open-weight 754-billion-parameter Mixture-of-Experts model from Z.AI that achieves state-of-the-art results on SWE-Bench Pro and can sustain autonomous agentic execution for up to 8 hours on a single task.
Read definitionGLM-5.2
GLM-5.2 is Z.AI's open-weight (~753B-parameter Mixture-of-Experts) flagship, released June 2026 under the MIT License. It pairs a genuinely usable 1-million-token context window with best-in-class open agentic-coding performance — 62.1 on SWE-bench Pro and 81.0 on Terminal-Bench 2.1 — at roughly one-sixth the price of comparable frontier models.
Read definitionGPT-OSS Series
OpenAI's first open-weight reasoning model family gpt-oss-120b and gpt-oss-20b released under Apache 2.0, bringing frontier-level intelligence to self-hosted deployments.
Read definitionStay ahead of the curve
Weekly newsletter on agentic AI, LLMs, and what we're building at Superteams — straight to your inbox.
Ready to ship AI in production?
We deploy fractional AI teams that deliver production-grade systems in 30–90 days. No fluff, no obligation.