filmov
tv
Qwen2.5 coder - Combines code generation with reasoning to build coding agents!

Показать описание
Qwen 2.5 coder is one of the specialized models released along with the general-purpose Qwen 2.5 LLM. The specialty of the model is coupling reasoning with code generation ability. This aims to create a coding agent rather than a simple code generation model.
So in this video, let's look at what makes Qwen 2.5 coder special, from systematically mixing the data in the right proportion to training.
If you wish to learn about the world's best code generation model to date, this video is for you!
⌚️ ⌚️ ⌚️ TIMESTAMPS ⌚️ ⌚️ ⌚️
0:00 - Intro
1:45 - Comparison to DeepSeek and CodeStral
2:25 - From CodeQwen to Qwen Coder
3:20 - Qwen 2.5 Base Models
4:40 - Model Architecture
5:43 - Model Training Data
8:59 - Data Mixture
10:16 - Model Training Policy
11:45 - Instruction Tuned Models
13:42 - Best performance-to-size ratio
14:00 - Math reasoning
15:05 - Extro
QWEN 2.5 CODER KEY LINKS
MY KEY LINKS
WHO AM I?
I am a Machine Learning researcher/practitioner who has seen the grind of academia and start-ups. I started my career as a software engineer 15 years ago. Because of my love for Mathematics (coupled with a glimmer of luck), I graduated with a Master's in Computer Vision and Robotics in 2016 when the AI revolution started. Life has changed for the better ever since.
#machinelearning #deeplearning #aibites
So in this video, let's look at what makes Qwen 2.5 coder special, from systematically mixing the data in the right proportion to training.
If you wish to learn about the world's best code generation model to date, this video is for you!
⌚️ ⌚️ ⌚️ TIMESTAMPS ⌚️ ⌚️ ⌚️
0:00 - Intro
1:45 - Comparison to DeepSeek and CodeStral
2:25 - From CodeQwen to Qwen Coder
3:20 - Qwen 2.5 Base Models
4:40 - Model Architecture
5:43 - Model Training Data
8:59 - Data Mixture
10:16 - Model Training Policy
11:45 - Instruction Tuned Models
13:42 - Best performance-to-size ratio
14:00 - Math reasoning
15:05 - Extro
QWEN 2.5 CODER KEY LINKS
MY KEY LINKS
WHO AM I?
I am a Machine Learning researcher/practitioner who has seen the grind of academia and start-ups. I started my career as a software engineer 15 years ago. Because of my love for Mathematics (coupled with a glimmer of luck), I graduated with a Master's in Computer Vision and Robotics in 2016 when the AI revolution started. Life has changed for the better ever since.
#machinelearning #deeplearning #aibites
Комментарии