--- title: EP231: AI Terms You Need to Know | daily.dev description: This newsletter edition is a glossary and explainer roundup covering AI terminology, GPT model history, AI/ML interview topics, and GPU architecture. It... canonical: https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: EP231: AI Terms You Need to Know | daily.dev og:description: This newsletter edition is a glossary and explainer roundup covering AI terminology, GPT model history, AI/ML interview topics, and GPU architecture. It... og:url: https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr og:image: https://api.daily.dev/og/posts/xzWMJHilr.png og:image:alt: EP231: AI Terms You Need to Know og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # EP231: AI Terms You Need to Know **[ByteByteGo](https://daily.dev/sources/bytebytego)** · 6 min read · 1 upvotes · 0 comments ## Summary This newsletter edition is a glossary and explainer roundup covering AI terminology, GPT model history, AI/ML interview topics, and GPU architecture. It defines agentic AI, large reasoning models, vector databases, RAG, and MCP in plain terms. It traces GPT-2 through GPT-6, noting GPT-3's 175B parameters, GPT-4's 32K context and RLHF tuning, GPT-5's 400K context and reasoning mode, and GPT-6's 1.05M context with async tool calling and mid-turn steering. It also breaks down GPU internals, including Streaming Multiprocessors, warp schedulers, CUDA cores, and the L1/L2 cache hierarchy, and lists traditional vs modern AI/ML interview topics. The issue also promotes Railway's cloud platform and an AI cost optimization course offered with a 20% discount for subscribers. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### How many parameters does GPT-3 have compared to GPT-2? GPT-3 has 175 billion parameters, more than 100 times the size of GPT-2's 1.5 billion parameters. Both models used the same pre-training recipe with next-token prediction, but GPT-3's much larger scale gave it in-context learning, letting it follow patterns from examples shown in a prompt, something GPT-2 could not do. _Comparing model generations like this helps you follow AI progress on daily.dev as new versions ship._ ### What new features does GPT-6 add over GPT-5? GPT-6 expands the context window to 1.05 million tokens, up from GPT-5's 400,000, and adds async tool calling along with a feature called mid-turn steering, letting the model take a new instruction while it is still thinking. It is also described as generally safer than prior generations. _Developers tracking model capability jumps like this stay current on daily.dev._ ### What is the difference between a Streaming Multiprocessor, CUDA cores, and L1/L2 cache in a GPU? A Streaming Multiprocessor (SM) is a mini parallel engine replicated across a GPU chip, and dozens of these work simultaneously instead of one central processor. Inside each SM, a warp scheduler assigns threads to run, CUDA cores execute instructions in parallel, and L1 cache gives fast on-SM data access. L2 cache sits below that, shared across all SMs, checked when an SM misses in L1 before going to slower, higher-bandwidth global memory. _Understanding GPU internals like this helps when choosing hardware for AI workloads, a topic you can follow on daily.dev._ ## Similar posts on daily.dev - [EP202: MCP vs RAG vs AI Agents](https://daily.dev/posts/ep202-mcp-vs-rag-vs-ai-agents-jowyr3vcd) · ByteByteGo · 5 upvotes · 0 comments - [The only AI glossary you'll need this year](https://daily.dev/posts/the-only-ai-glossary-you-ll-need-this-year-mewy6ddxs) · TechCrunch · 0 upvotes · 0 comments - [AI Crash Course: AI, ML, LLM and more](https://daily.dev/posts/ai-crash-course-ai-ml-llm-and-more-dduqcdz02) · Telerik · 3 upvotes · 0 comments - [30 Must-Know Agentic AI Terms](https://daily.dev/posts/30-must-know-agentic-ai-terms-ojlcnutal) · Daily Dose of Data Science \| Avi Chawla \| Substack · 1 upvotes · 0 comments --- Tags: [#ai](https://daily.dev/tags/ai), [#gpu](https://daily.dev/tags/gpu), [#rag](https://daily.dev/tags/rag), [#gpt](https://daily.dev/tags/gpt), [#vector-search](https://daily.dev/tags/vector-search) [View this post on daily.dev](https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"EP231: AI Terms You Need to Know","url":"https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr"},"datePublished":"2026-10-31T15:34:21.759Z","dateModified":"2026-10-31T15:34:52.952Z","description":"This newsletter edition is a glossary and explainer roundup covering AI terminology, GPT model history, AI/ML interview topics, and GPU architecture. It...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/0e9d1a7a11c419edd539bd20afc52381?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/0e9d1a7a11c419edd539bd20afc52381?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"ByteByteGo","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"ByteByteGo","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/35be29234ee14d01a9cd049c52e12753","url":"https://daily.dev/sources/bytebytego"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai,gpu,rag,gpt,vector-search","timeRequired":"PT6M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"ByteByteGo","item":"https://daily.dev/sources/bytebytego"},{"@type":"ListItem","position":3,"name":"EP231: AI Terms You Need to Know"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/ep231-ai-terms-you-need-to-know-xzwmjhilr#faq","mainEntity":[{"@type":"Question","name":"How many parameters does GPT-3 have compared to GPT-2?","acceptedAnswer":{"@type":"Answer","text":"GPT-3 has 175 billion parameters, more than 100 times the size of GPT-2's 1.5 billion parameters. Both models used the same pre-training recipe with next-token prediction, but GPT-3's much larger scale gave it in-context learning, letting it follow patterns from examples shown in a prompt, something GPT-2 could not do. Comparing model generations like this helps you follow AI progress on daily.dev as new versions ship."}},{"@type":"Question","name":"What new features does GPT-6 add over GPT-5?","acceptedAnswer":{"@type":"Answer","text":"GPT-6 expands the context window to 1.05 million tokens, up from GPT-5's 400,000, and adds async tool calling along with a feature called mid-turn steering, letting the model take a new instruction while it is still thinking. It is also described as generally safer than prior generations. Developers tracking model capability jumps like this stay current on daily.dev."}},{"@type":"Question","name":"What is the difference between a Streaming Multiprocessor, CUDA cores, and L1/L2 cache in a GPU?","acceptedAnswer":{"@type":"Answer","text":"A Streaming Multiprocessor (SM) is a mini parallel engine replicated across a GPU chip, and dozens of these work simultaneously instead of one central processor. Inside each SM, a warp scheduler assigns threads to run, CUDA cores execute instructions in parallel, and L1 cache gives fast on-SM data access. L2 cache sits below that, shared across all SMs, checked when an SM misses in L1 before going to slower, higher-bandwidth global memory. Understanding GPU internals like this helps when choosing hardware for AI workloads, a topic you can follow on daily.dev."}}]} ```