# stitionai/devika

Devika is the first open-source implementation of an Agentic Software Engineer. Initially started as an open-source alternative to Devin.

Repository: https://github.com/stitionai/devika
Canonical: https://ross.abutalabs.com/products/devika
Homepage: https://winfunc.com
Language: Python
License: MIT
License Family: permissive
Last push: 2025-09-25T08:47:36+00:00

## Health v2 (maintenance only)
Score: 44/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 43, release rhythm 35, longevity 63
- inputs: {"age_days": 895, "days_push": 342, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 19561, forks 2581 (observed 2026-08-28T04:11:28.873852+00:00)

## What it is
Devika is an open-source agentic AI software engineer that takes high-level instructions, plans steps, researches via web browsing, and writes code across multiple languages. It supports Claude, GPT-4, Gemini, Mistral, Groq, and local LLMs via Ollama, with a chat interface and project-based management.

## Use cases
- automate coding tasks from natural language instructions
- open-source alternative to Devin AI software engineer
- AI agent that researches and writes code autonomously
- break down a feature request into steps and implement it
- run a local LLM coding agent with Ollama
- AI pair programmer for building projects from scratch

## When to choose
- you want a self-hosted, open-source autonomous coding agent
- you want to experiment with agentic software engineering and contribute to an early-stage project
- you want to use local models via Ollama for code generation

## When to avoid
- you need a production-ready, reliable coding assistant - the project is explicitly experimental with broken features
- you want a polished commercial product with support
- you need guaranteed SWE-bench-level performance today

## Facets
- artifact type: application
- maturity: experimental
- function: agent-framework, llm-inference, chat-interface, web-scraping, developer-tools
- domain: large-language-models, developer-tools
- platform: python, self-hosted, cross-platform
- tags: ai-software-engineer, open-source-devin-alternative, autonomous-coding-agent, llm-agent, swe-bench, ollama, chat-ui, ai-agents, automation, nodejs, docker

## Member repositories
- stitionai/devika (main) score 44

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:28.873852+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:00:05.010375+00:00, confidence not recorded.
  - readme: https://github.com/stitionai/devika (fetched 2026-08-28T04:11:28.873852+00:00, sha 4beee0e9ff44)
  - homepage: https://winfunc.com (fetched 2026-08-29T07:58:30.282599+00:00, sha efa4aa1322d7)
  - site_page: https://winfunc.com/features (fetched 2026-08-29T07:58:30.298810+00:00, sha 41f8f182e6ec)
  - site_page: https://winfunc.com/about (fetched 2026-08-29T07:58:30.303802+00:00, sha 2783bd020b13)
  - site_page: https://winfunc.com/pricing (fetched 2026-08-29T07:58:30.296729+00:00, sha 7311ff8acdbb)
  - site_page: https://winfunc.com/integrations (fetched 2026-08-29T07:58:30.300480+00:00, sha c01443b642d7)
  - site_page: https://winfunc.com/changelog (fetched 2026-08-29T07:58:30.302135+00:00, sha e7f74a45a91c)
- Data as of 2026-08-30T08:39:29.467469+00:00.