CompactAI commited on
Commit
615bf9f
·
verified ·
1 Parent(s): ea40c7e

Upload 4 files

Browse files
Files changed (1) hide show
  1. index.html +4 -2
index.html CHANGED
@@ -396,7 +396,7 @@ blockquote { border-left: 4px solid var(--color-accent); padding-left: 1.5rem; m
396
  <h2 class="section-title">The Confession</h2>
397
  <div class="confession-content">
398
  <div class="confession-text">
399
- <p class="drop-cap">Let me be honest with you. For months, I was that person. You know the one. Leaving my computer on overnight, running distillation scripts, trying to squeeze GPT-4 into something that could run on a potato.</p>
400
 
401
  <p>And you know what? <strong>It was boring.</strong></p>
402
 
@@ -829,9 +829,11 @@ blockquote { border-left: 4px solid var(--color-accent); padding-left: 1.5rem; m
829
  <section class="closing" id="closing">
830
  <div class="container">
831
  <div class="closing-content">
832
- <h2>Beyond Competing with GPT-4</h2>
833
  <p>The real goal is understanding what's actually necessary for intelligence to emerge. Maybe 100K parameters is enough. Maybe it isn't. But we won't know until we try building from first principles instead of just compressing what already exists.</p>
834
 
 
 
835
  <div class="closing-quote">
836
  <blockquote>
837
  "Smaller models serve as a means toward a larger goal. Understanding what makes models work in the first place is the real objective."
 
396
  <h2 class="section-title">The Confession</h2>
397
  <div class="confession-content">
398
  <div class="confession-text">
399
+ <p class="drop-cap">Let me be honest with you. For months, I was that person. You know the one. Leaving my computer on overnight, running distillation scripts, trying to squeeze Claude opus 4.6 into something that could run on a potato.</p>
400
 
401
  <p>And you know what? <strong>It was boring.</strong></p>
402
 
 
829
  <section class="closing" id="closing">
830
  <div class="container">
831
  <div class="closing-content">
832
+ <h2>Beyond Competing with Claude opus 4.6</h2>
833
  <p>The real goal is understanding what's actually necessary for intelligence to emerge. Maybe 100K parameters is enough. Maybe it isn't. But we won't know until we try building from first principles instead of just compressing what already exists.</p>
834
 
835
+ <p><strong>FMN-GPT will be 100% open-source.</strong> This includes the training script, full model weights, all checkpoints, the base model, the instruction model, and our custom inference engine. We believe that open research accelerates progress for everyone.</p>
836
+
837
  <div class="closing-quote">
838
  <blockquote>
839
  "Smaller models serve as a means toward a larger goal. Understanding what makes models work in the first place is the real objective."