Devika

Devika

Open Source(MIT)
Verified
AI-Native
Superseded by a newer product (2025) — this version is now maintenance-only.

Open-source agentic AI software engineer (open Devin alternative)

Global
Self-hostable
Audit Logs
Multi-agent
Persistent memory
+1
Copilot
Last updated: Oct 13, 2026
Open Source
Visit Site19.6k
More

Quick Answer

Devika is an open-source agentic AI software engineer that turns high-level instructions into steps, researches the web, and writes code. It supports Claude 3, GPT-4, Gemini, Mistral, Groq and local LLMs via Ollama, with a browser-based project UI. Free and MIT-licensed, but explicitly in an early experimental stage.

About Devika

Devika is an open-source agentic AI software engineer that understands high-level instructions, breaks them into steps, researches the relevant information, and writes code to reach the objective. It offers planning and reasoning, web browsing for context, multi-language code generation, and project organisation through a chat interface. MIT-licensed, supporting Claude, GPT-4, Gemini, Mistral, Groq and local models via Ollama.

AI-extracted from public sources. May contain errors. Methodology · Report an error
What Makes Devika Unique

Devika started as an open-source alternative to Devin, aiming to match Devin's SWE-bench score, and combines AI planning, contextual keyword extraction for research and web browsing to write code toward an objective.

Decision Intelligence

Best For

Developers experimenting with autonomous coding agents on personal projects
Users who want to run an agentic coder locally against Ollama models
Contributors wanting to help build an open Devin alternative
Developers who need an agent that browses the web while writing code
Pricing Overview

Type

Open Source

Dimension Scores

· 3.9 avg

How good it is, across universal quality axes

Ease of Use1/10
Customization4/10
Support8/10
Integration1/10
Compliance & Data Protection1/10
Performance5/10
Planning Quality7/10
AI-scored from up to 37 public sources, reviewed by a person·High confidenceHow we score

Task Scores

· 3.4 avg

What it can do, per category it works in · showing 1 of 9 tasks scoring 5+

Multi Language Support6/10

Scores are AI-estimated from publicly available data — not an independent test or a verified user rating. How we rank →

Capabilities
Code
Code Review
Automation
Categories
Code Assistant
Developer Platform
Project Management
Limitations
  • -Explicitly early-stage: 'There are a lot of unimplemented/broken features at the moment'
  • -Requires Python 3.10 to 3.11, Node.js 18+, bun and uv
  • -Requires your own API keys for search (Bing/Google) and for each model provider
  • -Best performance requires the Claude 3 model family per the README's own note
Fit Score

Complexity

Advanced

Company Size

Solo
Small
Medium
Enterprise

Team Size

Individual
Small Team
Large Team

Skill Level

Intermediate
Expert
Features
Supports Claude 3, GPT-4, Gemini, Mistral, Groq and local LLMs via Ollama
Advanced AI planning and reasoning capabilities
Contextual keyword extraction for focused research
Seamless web browsing and information gathering
Code writing in multiple programming languages
Dynamic agent state tracking and visualization
Natural language interaction via chat interface
Project-based organization and management
Technical Capabilities
SDK Available
MCP Compatible (unknown)
Multi-Agent
Human-in-Loop (unknown)
Team Features
Webhook Support (unknown)
Has Demo
Function Calling (unknown)
Sandbox Mode (unknown)
Audit Logs
No Signup (unknown)
Works Offline (unknown)
Requires Host App
Framework
MCP Server
Digital Employee

Deployment

Self Hosted
Compliance, Privacy & Trust
GDPR: No info available
SOC 2: No info available
HIPAA: No info available
Trains on user data: Unknown
Human review:
Recommended
Tech Specs
Response time (est.):
Moderate

The Model(s) Behind It

Model details — extracted from public vendor sources by our pipeline (AI extraction plus a source check) or admin-entered. Shown as claimed by the vendor; not independently audited by Intelloro.

Powered By

Base Models

Training Data Provenance
Not disclosed
Relevant to EU AI Act Art. 53 / Annex XI (GPAI)
Supported Tools
Web Browser
Web Search
Code Interpreter
Terminal
File System
Pricing Plans
Free to use.

Looking for more Devika alternatives?

Explore our complete list of alternatives with filters for free, cheaper, and higher-rated options.

View All Alternatives

User Reviews

No reviews yet. Be the first to review!

Who It's For
4 roles
Software DevelopersComputer ProgrammersWeb Developers+1 more
Platform, Content & Languages
4 content types
1 languages
CodeTextDocumentEnglish+9 more

Content Types

Code
Text
Document
Spreadsheet

Accepts

Code
Text
Document
API

Produces

Code
Text
Document
API
Languages: English

Frequently Asked Questions

What can Devika do?

Devika can: Supports Claude 3, GPT-4, Gemini, Mistral, Groq and local LLMs via Ollama, Advanced AI planning and reasoning capabilities, Contextual keyword extraction for focused research, Seamless web browsing and information gathering. Best for: Developers experimenting with autonomous coding agents on personal projects.

How much does Devika cost?

Devika is open source and free to self-host. Cloud hosting options may have separate pricing.

Is Devika free?

Yes, Devika is open source and free to self-host. Managed hosting may carry separate costs.

What makes Devika unique?

Devika started as an open-source alternative to Devin, aiming to match Devin's SWE-bench score, and combines AI planning, contextual keyword extraction for research and web browsing to write code toward an objective.

What platforms does Devika support?

Devika is available on web and CLI.

What languages does Devika support?

Devika supports English.

Who is Devika for?

Devika is designed for Developers experimenting with autonomous coding agents on personal projects; Users who want to run an agentic coder locally against Ollama models; Contributors wanting to help build an open Devin alternative.

Trust Score

60/100

Good

Based on 5 data signals

Confidence: High

Reflects publicly available data, not an independent audit

Compliance & Verified38/38
VerifiedYes
Open-source licenseMIT
Human ReviewYes
Reliability0/10

No published data

Company17/17
Company ListedYes
Team Size1-10
Health—

Not yet checked

AI Governance & Regulatory Disclosures

Legally-sensitive fields — extracted from public vendor sources by our pipeline (AI extraction plus a source check) or admin-entered. Shown as claimed by the vendor; not independently audited by Intelloro.

EU AI Act Risk Tier
Unclassified

Switching & Migration

Migration Difficulty
Moderate Effort
Support:
Discord
GitHub
Quick Stats
Agent TypeCopilot
MemoryLong Term
PlanningAdvanced
Base ModelGPT-4 +1
PlatformsWeb, CLI
Launched2024
VerifiedOct 2026
Company & Links
Company Namewinfunc (formerly Stition AI)
Founded2024
Based inIndia
Founders
Documentation