Créer des GPTs sans abonnement ChatGPT - Cette IA est FOLLE !

Créer des GPTs sans abonnement ChatGPT - Cette IA est FOLLE !

by Ludo Salenne

Applied Sciences & Engineering AI — Models & Research UYQArtificial intelligence (AI)UYQFGenerative artificial intelligence
🎙 Ludo Salenne 👥 267K 📅 February 5, 2024 ⏱ 23 min 👁 21K 📄 tutorial 🧭 2026-08-21
Available in: English (current) Français

Keywords

HuggingChatGPTsMistral AIOpen SourceTutorial

Summary

The video presents HuggingChat’s new feature that allows users to create and use GPTs (called assistants) for free, without a ChatGPT subscription. The creator demonstrates how to access the assistant store, filter by model (e.g., Mistral, Llama), and create a custom assistant by providing a name, description, and instructions. He then tests the assistant with a productivity coaching scenario, showing how it uses the Eisenhower matrix to prioritize tasks. The video also compares the performance of a complex GPT (Master AI) on ChatGPT vs. HuggingChat, highlighting the speed and responsiveness of the latter. Additionally, he adapts his ‘Pimp My Prompt’ prompt to HuggingChat, showing how it generates a detailed prompt for a business document. The creator emphasizes the advantages of HuggingChat: free, open-source, no message limits, and access to models like Mistral. He concludes by encouraging viewers to try the feature and mentions that assistants are publicly shared on the store.

151 words

Critical Evaluation

Value of the Information & Strength of the Argument

The video provides practical, hands-on demonstrations of creating and using GPTs on HuggingChat, which is valuable for users seeking free alternatives to ChatGPT. The argumentation is based on direct comparisons between ChatGPT and HuggingChat, showing the latter’s speed and effectiveness in executing complex prompts. However, the claims about performance are anecdotal and lack quantitative benchmarks. The creator’s enthusiasm is evident, but the argumentation could be strengthened by more systematic testing and references to official documentation.

Scientific Rigor, Source Quality, Title Accuracy

The video does not cite external scientific sources, but it references the HuggingChat platform and the creator’s own resources. The title accurately reflects the content, which is a tutorial on creating GPTs without a subscription. The video is well-structured with clear chapters, but the lack of citations and the subjective nature of the performance comparisons reduce its scientific rigor. The description includes links to the HuggingChat assistant page and the creator’s other videos, which serve as additional resources.

168 words

Title / Content Match

The title accurately reflects the content: the video demonstrates how to create and use GPTs without a ChatGPT subscription, using HuggingChat.

Quality & Reliability

6/10

The video is a practical tutorial demonstrating the creation and use of GPTs on HuggingChat. It is based on direct experience and demonstrations, but lacks in-depth technical explanations or citations of official documentation. The claims about performance are subjective and not backed by benchmarks.

Chapters

Cited Sources

Concurring Sources

Contribution & Novelties

The video provides a timely tutorial on a new feature of HuggingChat, offering a free and open-source alternative to ChatGPT’s GPTs. It demonstrates the practical steps to create custom assistants and highlights the performance of Mistral models. The comparison with ChatGPT’s GPTs is useful for users considering alternatives.

Pour aller plus loin :

  • Hugging Face — The platform behind HuggingChat, central to the open-source AI ecosystem.
  • Mistral AI — The French AI company providing the models used in the demonstrations.
  • Llama (Meta AI) — Meta’s open-source language models, also available on HuggingChat.
  • Eisenhower Matrix — A productivity framework referenced in the video for task prioritization.

105 words

Radar Profile

The radar chart shows a balanced profile with moderate scores across all dimensions. The video excels in practical demonstration (quantité d'information) but lacks depth in technical explanation and scientific rigor, resulting in a moderate overall quality.

Reliability 6/10