Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
41 tokens/sec
GPT-4o
59 tokens/sec
Gemini 2.5 Pro Pro
41 tokens/sec
o3 Pro
7 tokens/sec
GPT-4.1 Pro
50 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

CodeGemma: Open Code Models Based on Gemma (2406.11409v2)

Published 17 Jun 2024 in cs.CL and cs.AI

Abstract: This paper introduces CodeGemma, a collection of specialized open code models built on top of Gemma, capable of a variety of code and natural language generation tasks. We release three model variants. CodeGemma 7B pretrained (PT) and instruction-tuned (IT) variants have remarkably resilient natural language understanding, excel in mathematical reasoning, and match code capabilities of other open models. CodeGemma 2B is a state-of-the-art code completion model designed for fast code infilling and open-ended generation in latency-sensitive settings.

User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (27)
  1. CodeGemma Team (1 paper)
  2. Heri Zhao (1 paper)
  3. Jeffrey Hui (6 papers)
  4. Joshua Howland (5 papers)
  5. Nam Nguyen (46 papers)
  6. Siqi Zuo (3 papers)
  7. Andrea Hu (3 papers)
  8. Christopher A. Choquette-Choo (49 papers)
  9. Jingyue Shen (3 papers)
  10. Joe Kelley (2 papers)
  11. Kshitij Bansal (12 papers)
  12. Luke Vilnis (20 papers)
  13. Mateo Wirth (10 papers)
  14. Paul Michel (27 papers)
  15. Peter Choy (4 papers)
  16. Pratik Joshi (7 papers)
  17. Ravin Kumar (10 papers)
  18. Sarmad Hashmi (1 paper)
  19. Shubham Agrawal (21 papers)
  20. Zhitao Gong (12 papers)
Citations (64)
X Twitter Logo Streamline Icon: https://streamlinehq.com