An open-source model based on GLM-4.7-Flash, optimized specifically for code generation and development workflows.

tools thinking

ollama run DedeProGames/codex-oss

curl http://localhost:11434/api/chat \
  -d '{
    "model": "DedeProGames/codex-oss",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='DedeProGames/codex-oss',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'DedeProGames/codex-oss',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 6 days ago

6 days ago

62428eb62166 · 19GB ·

model

archglm4moelite

parameters29.9B

quantizationQ4_K_M

19GB

license

Could not load data.

params

{ "num_ctx": 32768, "temperature": 1 }

34B

template

13B

Readme

Welcome to codex-oss

Welcome to codex-oss, my open-source model based on GLM-4.7-Flash, optimized specifically for code generation and development workflows.

Built to deliver high efficiency, high speed, and strong performance across any type of project, codex-oss is designed to maximize productivity while maintaining lightweight execution.

codex-oss aims to provide a powerful yet streamlined coding experience.