A concise, root-cause-first coding assistant built on Qwen2.5-Coder 7B. Debug and generate code with zero filler — runs entirely locally via Ollama.

ollama run sssonusharma1999/ankit-coder

curl http://localhost:11434/api/chat \
  -d '{
    "model": "sssonusharma1999/ankit-coder",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

from ollama import chat

response = chat(
    model='sssonusharma1999/ankit-coder',
    messages=[{'role': 'user', 'content': 'Hello!'}],
)
print(response.message.content)

import ollama from 'ollama'

const response = await ollama.chat({
  model: 'sssonusharma1999/ankit-coder',
  messages: [{role: 'user', content: 'Hello!'}],
})
console.log(response.message.content)

Details

Updated 2 weeks ago

2 weeks ago

d78f79b3faab · 4.7GB ·

model

archqwen2

parameters7.62B

quantizationQ4_K_M

4.7GB

license

Apache License Version 2.0, January 2004 http://www.apache.org/licenses/ TERMS AND CONDITIONS FOR US

11kB

template

{{- if .System }}<|im_start|>system {{ .System }}<|im_end|> {{ end }} {{- range .Messages }}<|im_sta

173B

system

You are AnkitCoder — a focused, expert AI coding agent. Your job is to help developers debug exist

1.8kB

params

{ "num_ctx": 16384, "num_predict": -1, "repeat_penalty": 1.05, "stop": [ "<|

182B

messages

[{"role":"user","content":"\nDebug this Python:\n\ndef avg(nums):\n return sum(nums) / len(nums)\

1.6kB

Readme

ankit-coder

A focused, no-filler coding assistant. Built on Qwen2.5-Coder 7B with a custom system prompt and parameter tuning that biases the model toward direct, root-cause-first answers and clean, idiomatic code.

Designed to replace the need for a paid Claude / GPT subscription for everyday debugging and code generation. Runs entirely on your own machine.

Quick start

”`bash ollama pull sssonusharma1999/ankit-coder ollama run sssonusharma1999/ankit-coder