codybum commited on
Commit
265c752
·
verified ·
1 Parent(s): 5c918b6

Rebrand to KOS-V4; David-vs-Goliath GPT-3.5 framing; complete peer tables

Browse files
Files changed (1) hide show
  1. README.md +1 -1
README.md CHANGED
@@ -36,7 +36,7 @@ data and compute.
36
 
37
  **And it reaches that bar.** On **IFEval**, the standard instruction-following benchmark, KOS-V4-Instruct scores
38
  **AVG4 62.7**, on par with and a touch above the reported **GPT-3.5-turbo (60.7)**, and within about three points of
39
- a current big-company 3B, **Qwen2.5-3B-Instruct (65.4)**.
40
 
41
  | IFEval (AVG4 of the four strict and loose metrics) | score | who built it, and how |
42
  | :-- | :-- | :-- |
 
36
 
37
  **And it reaches that bar.** On **IFEval**, the standard instruction-following benchmark, KOS-V4-Instruct scores
38
  **AVG4 62.7**, on par with and a touch above the reported **GPT-3.5-turbo (60.7)**, and within about three points of
39
+ a commercial 3B, **Qwen2.5-3B-Instruct (65.4)**.
40
 
41
  | IFEval (AVG4 of the four strict and loose metrics) | score | who built it, and how |
42
  | :-- | :-- | :-- |