bigeagle commited on
Commit
dc514c7
·
verified ·
1 Parent(s): c5d1dd4

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -1
README.md CHANGED
@@ -29,10 +29,11 @@ pipeline_tag: image-text-to-text
29
 
30
 
31
  <p align="center">
32
- 📰&nbsp;&nbsp;<a href="https://www.kimi.com/blog/kimi-k3">Tech Blog</a>
33
  </p>
34
 
35
 
 
36
  ## 1. Model Introduction
37
 
38
  Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date. It is a 2.8T-parameter model built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes), with native vision capabilities and a 1-million-token context window. It is the world's first open 3T-class model, designed for frontier intelligence across long-horizon coding, knowledge work, and reasoning.
 
29
 
30
 
31
  <p align="center">
32
+ 📰&nbsp;&nbsp;<a href="https://www.kimi.com/blog/kimi-k3">Tech Blog</a> | &nbsp;&nbsp;&nbsp; <b>📄&nbsp;&nbsp;<a href="k3_tech_report.pdf">Full Report</a></b>
33
  </p>
34
 
35
 
36
+
37
  ## 1. Model Introduction
38
 
39
  Kimi K3 is an open-weight, native multimodal agentic model and our most capable model to date. It is a 2.8T-parameter model built on Kimi Delta Attention (KDA) and Attention Residuals (AttnRes), with native vision capabilities and a 1-million-token context window. It is the world's first open 3T-class model, designed for frontier intelligence across long-horizon coding, knowledge work, and reasoning.