Compare AI tools

	GitNexus A	Ollama S	Perplexity S	Le Chat (Mistral) B
Tagline	An open-source, MCP-native knowledge graph engine that gives AI coding agents (Cursor, Claude Code, Windsurf) genuine structural awareness of your codebase before they touch a single line.	Run LLMs locally. One-line install, GUI optional.	AI search done right. Cited answers, not chat theater.	French alternative. Fast, European, privacy-focused.
Category	Coding	Dev Platform	Research	Chatbots
Pricing	Free (MIT open source)	Free + open source	Free + $20/mo Pro	Free + $15/mo Pro
Best for	Developers working in large or unfamiliar codebases who want their AI coding agent to stop making confident, structurally blind edits — especially Claude Code power users.	Devs wanting offline/local LLMs for privacy or experimentation.	Replacing Google for any question where you want a cited answer in seconds.	European users with data residency needs. Fans of open-weight models.
Strengths	Pre-computes a full dependency graph (functions, imports, class inheritance, execution flows) via Tree-sitter ASTs — agents query structure, they don't guess at it Zero-server, privacy-first: CLI runs entirely locally with no network calls; browser UI processes code client-side and never uploads it Deepest Claude Code integration on the market: MCP tools + agent skills + PreToolUse/PostToolUse hooks that auto-enrich searches and auto-reindex after commits One global MCP server handles multiple indexed repos — set up once with npx gitnexus setup and forget it detect_impact and generate_map MCP prompts give pre-commit blast-radius analysis and auto-generated Mermaid architecture docs	Run Llama, Mistral, Qwen, etc. on your laptop Simple CLI + API Hardware-aware (picks the right quant)	Sources every claim Fast, current answers Pro Search runs multi-step research Spaces for persistent context	European data residency Very fast responses Open-weight Mistral models available Good French/European languages
Weaknesses	Browser-side RAG has hard ceilings: WASM heap limits constrain embedding model quality compared to server-side tools; monorepos or repos >50k files hit practical walls Community-built and not officially maintained — velocity and long-term support depend on contributor goodwill Claude Code gets the full integration experience; other editors (Windsurf, Cursor) get progressively less — value is uneven depending on your editor	Needs beefy laptop for larger models Speed way behind cloud APIs	Not a general chatbot Answers can be shallow on complex topics	Smaller capability gap vs frontier models Less polished UX
Kai's verdict	GitNexus solves a real and underappreciated problem: AI coding agents are syntactically fluent but architecturally blind, and plugging a pre-computed knowledge graph into the MCP layer is the right fix. 28k GitHub stars in days suggests the pain is widely felt — just go in knowing it's a community project, not a polished product. (Verdict pending Phi's full review.)	S-tier for local inference. If you care about privacy or want to tinker, install this today.	S-tier for search. I use it before Google now. If you're still Googling everything, try this for a week.	B-tier overall, A-tier if GDPR/data residency matters. Solid backup option.
Link	Open →	Open →	Open →	Open →