Hello everyone!

I wanted to write this article to share my experience with agentic coding without Claude and Codex,
the most popular tools out there.

Introduction

I started dabbling with agentic coding a few months ago when Claude had decent limits on the 20$ plan,
and I was really impressed by the "fire and forget" way of doing things.

You prompt the agent: I want e2e tests, and it will study the codebase and implement them.

When I've started hitting limits on Claude code, and this is not a secret that they reduced limits for everyone on the
20$ subscription plan, I've moved to Codex because limits were more reasonable.

After a month of Codex, I've started exploring other tools.

OpenCode

OpenCode (https://opencode.ai/) is a OSS agentic coding tool similar to claude code and codex, and it has similar
features like plan mode, MCP server support, AGENTS.md, and SKILLs support.

The amazing thing about it is that it supports almost every AI Provider that's out there,
and it also supports running local LLMs.

One service that is offered by OpenCode directly is Zen service, which is basically
pay-as-you-go for curated models.

Deepseek V4 Pro

I've spent a weekend building a side project using DeepSeek V4 Pro using the pay as you go pricing.
The model is currently discounted at 75%, and I've used around 60M tokens and only paid 1.87$.

I could not sense any difference in quality compared to Claude or Codex. I gave it a vague prompt like implement
end-to-end testing with UI, and it explored the code base, generated the relevant tests, ran them,
iterated though them, and in a few minutes it added them.

Another feature that I've built with it is a page which allows the user to import transactions into the app from a
bank statement, it parses the statement and allows the user to edit the transaction before confirming the import.

This feature was implemented end to end and polished, after a few iterations.

Overall I'm quite satisfied by it and I don't feel the need to go back to Claude or Codex.

PI.dev

PI.dev (https://pi.dev/) is another OSS agentic coding tool. It's more minimalistic and OpenCode,
it doesn't have MCP support or "Plan Mode" and I don't have anything else to say about it because I haven't tried it.

Soulforge

Soulforge (https://github.com/proxysoul/soulforge) is quite different from the tools from above. It doesn't treat
the codebase as text, and it tries to build a knowledge graph. I haven't tried this one yet, but I wanted to mention
it because I thought it's interesting.

Conclusion

If you're unhappy with current providers you can try other tools too, other providers. It's not just Codex and Claude.
After coding for a whole weekend with DeepSeek V4 Pro and OpenCode I haven't sensed any difference,
once I got used to OpenCode it felt like I was still coding with Codex/Claude.

Thanks for reading!