ERNIE-4.5 0.3B First Look & LOCAL Test — Is a 300M LLM Any Good? скачать в хорошем качестве

ERNIE-4.5 0.3B First Look & LOCAL Test — Is a 300M LLM Any Good?

Timestamps: 00:00 - Intro 01:22 - Model Family First Look 02:10 - Technical Look 03:28 - Local Testing Setup 04:20 - HTML Site Test 07:32 - VRAM Usage 07:48 - Python Game Test 09:59 - Why Micro Models Matter 10:59 - Story Generation Test 12:53 - Refusal Test 13:56 - Roleplay Testing 15:48 - Closing Thoughts AI Integration Consulting: https://bijanbowen.com Join the Discord:   / discord   In this video, we take a look at the newly released ERNIE 4.5 family of models from Baidu. While the lineup includes models of various sizes, this video specifically focuses on the 0.3B parameter variant — a 300M LLM small enough to run nearly anywhere, but still capable of producing intelligent output. We test the model across several real-world tasks to evaluate its usability and limitations. This includes generating HTML websites, building simple Python games, performing story generation, and even roleplay-style interaction. We also briefly discuss the importance of micro models in the broader ecosystem and who this model might be useful for. Github Gist For Web Chat UI: https://gist.github.com/OminousIndust... HF Model Repo: https://huggingface.co/baidu/ERNIE-4....