Subscribe
Sign in
Home
Chat
š Start here!
𦄠Finetuning Sessions
šļø AI Systems Engineering
š Production OCR
Archive
About
Finetuning Sessions
Latest
Top
Discussions
From Finetuning to Deployment in One Live Session
The grand finale of the Finetuning Sessions!
Apr 6
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
24
1
2
56:33
Run the World's Best OCR on Your Own Laptop
Finetuning Sessions Ā· Lab 8 / 8
Apr 3
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
71
2
6
A Practical Guide to LLM Inference at Scale
Finetuning Sessions Ā· Lesson 8 / 8
Apr 1
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
96
5
12
The Builder's Guide to Multimodal Finetuning (Vision + TTS)
Finetuning Sessions Ā· Lab 7 / 8
Mar 27
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
24
4
Beyond Text: A Guide to Vision & TTS Finetuning
Finetuning Sessions Ā· Lesson 7 / 8
Mar 25
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
21
4
Finetuning Sessions ā Week 6 Office Hours
Reward Functions, Reward Hacking & GRPO Variants (DAPO, GSPO, Dr. GRPO)
Mar 23
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
11
1
1:08:33
Training a Reasoning Model on Consumer Hardware with GRPO and vLLM
Finetuning Sessions Ā· Lab 6 / 8
Mar 20
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
20
1
4
The RL Algorithm Behind DeepSeek's Reasoning Models
Finetuning Sessions Ā· Lesson 6 / 8
Mar 18
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
20
3
Finetuning Sessions ā Week 5 Office Hours
RL for LLM Alignment with PPO, DPO & KTO
Mar 16
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
6
1
45:24
KTO for Real-World LLM Alignment - A Hands-On Guide
Finetuning Sessions Ā· Lab 5 / 8
Mar 13
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
17
5
The RLHF Landscape - Aligning LLMs Beyond SFT
Finetuning Sessions Ā· Lesson 5 / 8
Mar 11
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
20
5
Finetuning Sessions ā Week 4 Office Hours
Covering QLoRA, GPU memory and quantization techniques
Mar 9
Ā
ā¢
Ā
Miguel Otero Pedrido
Ā andĀ
Antonio Zarauz Moreno
10
1
52:06
This site requires JavaScript to run correctly. Please
turn on JavaScript
or unblock scripts