← Back to OpenAI videos ← Terug naar OpenAI-video's
OpenAI VIDEO VIDEO 3 September 2025 3 september 2025
YouTube

Reinforcement fine-tuning (RFT) lets you improve how models reason by training with graders instead of large labeled datasets. This Build Hour shows you how to set up tasks, design...

Build Hour: Reinforcement Fine-Tuning Build Hour: Reinforcement Fine-Tuning

Reinforcement fine-tuning (RFT) lets you improve how models reason by training with graders instead of large labeled datasets. This Build Hour shows you how to set up tasks, design grading functions, and run efficient training loops with ju... Reinforcement fine-tuning (RFT) helpt je om het redeneervermogen van modellen te verbeteren door te trainen met graders in plaats van grote gelabelde datasets. Deze Build Hour laat zien hoe je taken opzet, beoordelingsfuncties ontwerpt en efficiënte trainingsloops draait met ju...

Video details Videogegevens
AI maker AI-maker OpenAI Published Gepubliceerd 3 September 2025 3 september 2025 Channel Kanaal OpenAI Playlist Playlist Uploads from OpenAI Updates Updates Videos Video's Watch on YouTube Bekijk op YouTube
Why it matters Waarom dit telt

Quick editorial signal Snelle redactionele duiding

1 min
Impact Impact

Relevant if you build with AI tools, APIs, or coding agents. Relevant als je bouwt met AI-tools, API's of coding agents.

Audience Voor wie Developers Developers
Level Niveau Expert Expert
  • Track this as a OpenAI update, not just a standalone headline. Bekijk dit als OpenAI-update, niet alleen als losse headline.
  • Useful for builders who need to understand API, coding, or workflow changes. Nuttig voor bouwers die API-, code- of workflowwijzigingen willen begrijpen.
  • Use the reactions below to tell us if this needs follow-up coverage. Gebruik de reacties hieronder om aan te geven of dit opvolging verdient.
model apps video developers

About this video Over deze video

Reinforcement fine-tuning (RFT) lets you improve how models reason by training with graders instead of large labeled datasets. This Build Hour shows you how to set up tasks, design grading functions, and run efficient training loops with just a few hundred examples.

Prashant Mital and Theophile Sautory (Applied AI) cover: - Intro to RFT: optimization, fine-tuning options, RFT benefits - Task setup: prompts, graders, and training and validation data - Live demo: building and running RFT for a classification task - RFT workflow: from dataset selection to evaluating and iterating - Customer spotlight: Accordance uses RFT models for tax and accounting workflows (https://accordance.com/) - Live Q&A

👉 Follow along with the code repo: https://github.com/openai/build-hours 👉 RFT Cookbook: https://cookbook.openai.com/examples/reinforcement_fine_tuning 👉 RFT Use Case Guide: https://platform.openai.com/docs/guides/rft-use-cases 👉 Sign up for upcoming live Build Hours: https://webinar.openai.com/buildhours

Help shape what we cover next Help bepalen wat we hierna volgen

Anonymous feedback, no frontend account needed. Anonieme feedback, zonder front-end account.

More videos from OpenAI Meer video's van OpenAI

All videos Alle video's
Ritu vs Case Files | With ChatGPT
OpenAI
27 Apr 2026 27 apr. 2026

Ritu vs Case Files | With ChatGPT Ritu vs Case Files | With ChatGPT

Space for focused work with ChatGPT Credits Director: Abhinav Pratiman DOP: Tassaduq Hussain Production House: Early Man Film Creative agency: Hue & Why Space for focused work with ChatGPT Credits Director: Abhinav Pratiman DOP: Tassaduq Hussain Production House: Early Man Film Creative agency: Hue & Why

Open video → Open video →
Reddys vs Retirement | With ChatGPT
OpenAI
25 Apr 2026 25 apr. 2026

Reddys vs Retirement | With ChatGPT Reddys vs. pensioen | Met ChatGPT

Start your second innings with ChatGPT Credits Director: Abhinav Pratiman DOP: Tassaduq Hussain Production House: Early Man Film Creative agency: Hue & Why Begin je tweede innings met ChatGPT Credits Regisseur: Abhinav Pratiman DOP: Tassaduq Hussain Productiehuis: Early Man Film Creatief bureau: Hue & Why

Open video → Open video →
Introducing GPT-5.5 with Perplexity
OpenAI
24 Apr 2026 24 apr. 2026

Introducing GPT-5.5 with Perplexity GPT-5.5 introduceren met Perplexity

“I thought building this internal tool was going to take me days…but with Codex and GPT-5.5, I was able to do it in under an hour.” Denis from Perplexity saw GPT-5.5 cut token usage by 56% while running the same agentic workflows faster an... “Ik dacht dat het bouwen van deze interne tool me dagen zou kosten… maar met Codex en GPT-5.5 kon ik het in minder dan een uur doen.” Denis van Perplexity zag dat GPT-5.5 het tokenverbruik met 56% verlaagde, terwijl dezelfde agentische workflows sneller en efficiënter draaiden...

Open video → Open video →
Workspace agents in ChatGPT: Weekly metrics reporting agent
OpenAI
24 Apr 2026 24 apr. 2026

Workspace agents in ChatGPT: Weekly metrics reporting agent Workspace-agents in ChatGPT: agent voor wekelijkse rapportage van statistieken

Watch a guided walkthrough of an agent that pulls Friday metrics, creates charts, drafts the narrative, and delivers a ready-to-share business report. Bekijk een begeleide rondleiding door een agent die vrijdagcijfers ophaalt, grafieken maakt, de tekst opstelt en een direct te delen bedrijfsrapport oplevert.

Open video → Open video →

Gemini komt eraan