Best AI coding tools that work with your own model
If you want to choose the model (or pay for it directly with your own API key), these coding tools are built to work across many model providers rather than one.
Ranked by score on our 100-point rubric (how we score). Scores are initial editorial estimates from public information, not hands-on tests. Pricing is labeled verified (checked on the vendor's official pricing page, with date) or unverified. Tools from more than one category appear here; each is scored within its own primary category.
Quick picks
| # | Tool | Tier | Score |
|---|---|---|---|
| 1 | Cursor AI-first code editor | S | 88 |
| 2 | Cline Open-source autonomous coding agent for VS Code | A | 80 |
| 3 | OpenCode Open-source terminal coding agent | A | 77 |
| 4 | Aider Terminal pair programmer with git integration | A | 73 |
| 5 | Dyad Local, open-source Lovable-style app builder | A | 70 |
The ranking
-
VS Code-based editor with tab completion, inline edits and background/cloud agents across many models.
- Best for
- You want an AI IDE that feels like VS Code with agents built in.
- Skip it if
- You prefer staying in JetBrains or the terminal.
- Pricing
- Free Hobby tier; paid Pro plan unverified editorial summary, not yet confirmed on an official pricing page
- Platforms
- Mac · Win · Linux · CLI
-
Open-source agent extension that plans and edits with your own model keys and supports MCP.
- Best for
- You want an open-source agent and full control of model choice and cost.
- Skip it if
- You want flat-rate pricing.
- Pricing
- Free and open source; pay your model provider unverified editorial summary, not yet confirmed on an official pricing page
- Platforms
- IDE extension · MCP
-
Terminal-first open-source agent with a polished TUI that works with 75+ model providers.
- Best for
- You want a Claude Code-style experience with any model.
- Skip it if
- You want a vendor-supported product.
- Pricing
- Free and open source; pay your model provider unverified editorial summary, not yet confirmed on an official pricing page
- Platforms
- CLI
-
Open-source CLI that edits code in your repo and auto-commits, working with most LLMs.
- Best for
- You like the terminal and clean git commits per change.
- Skip it if
- You want a GUI or a fully autonomous agent.
- Pricing
- Free and open source; pay your model provider unverified editorial summary, not yet confirmed on an official pricing page
- Platforms
- CLI
-
Free desktop app that builds full-stack apps locally using your own model keys or local models.
- Best for
- You want vibe coding without lock-in or cloud credits.
- Skip it if
- You want hosted deploys and zero setup.
- Pricing
- Free forever app; Dyad Pro $20/mo; Dyad Max $79/mo ✓ verified on the official pricing page, Oct 18, 2026
- Platforms
- Mac · Win · Linux
Head-to-head comparisons
FAQ
What is the best AI tool for coding with your own model or API key?
Cursor ranks first with 88/100 (S-tier). AI-first code editor. Use it when you want an AI IDE that feels like VS Code with agents built in. Next in our ranking: Cline (80/100, A-tier) and OpenCode (77/100, A-tier).
Which AI tools for coding with your own model or API key have a free tier?
In our pricing data, Cursor (unverified), Cline (unverified), OpenCode (unverified), Aider (unverified) and Dyad (verified) list a free tier. "Unverified" means it is an editorial summary we have not yet confirmed on the vendor's official pricing page, so check the vendor before relying on it.
Are there open-source options for coding with your own model or API key?
Yes. Cline, OpenCode, Aider and Dyad are listed as open source or open weights. Check each project's license before commercial use.
How are these tools ranked?
Each tool is scored out of 100 on our public rubric: job fit (30), ease (15), value (15), momentum (15), openness and portability (10), trust and maturity (10) and a builder/gamer bonus (5). S-tier is 85+, A-tier 70–84, B-tier 55–69 and C-tier below 55. This guide mixes tools from more than one category, and scores are calibrated within each tool's primary category, so treat cross-category gaps as rough. Current scores are initial editorial estimates from public information, not hands-on tests, and sponsors can't buy points.