OpenCode
Open-source, model-agnostic AI coding agent harness with terminal, web and desktop clients.
Liked as a customizable, model-agnostic open-source coding harness1234 that pairs well with cheap open models via the $10/month OpenCode Go plan or free models.5678 The most common complaints are about privacy and security, including a since-removed default of summarizing chats with a cloud model and unclear API key storage,910111213 followed by heavy RAM and CPU use.141516
Why people recommend it
Model flexibility · 3
Works with many providers, subscriptions, hosted APIs and local models.
“Since it's model agnostic, this might become my daily driver.”
“I use OpenCode and Pi connected to local models and to stuff like hosted DeepSeek all the time.”
Show 1 more
“I have been using OpenCode for most work related stuff now. It works well with my ChatGpt Pro subscription”
Cheap model access · 3
OpenCode Go and free models are seen as good value for heavy use.
“With my OpenCode Go sub of $10/month, DeepSeek is basically unlimited”
Show 1 more
“I use only OpenCode free models and occasional 20$ sub from OpenAi or Anthropic. And get complete/working products out of them.”
Customization · 3
Easy to extend with plugins and skills, or fork into your own harness.
“It is quite a synergy : my nix usage profits from the agent”
Show 1 more
“It's so easy to take opencode or pi and build my own for exactly what I want”
Daily driver · 3
Several commenters use it as their main coding tool, including for non-trivial apps.
“been using ocv2 as my daily driver for a while and had very little complaints. its a great product from a great team.”
“OC is all right -- it's my daily driver -- but it could be better.”
Show 1 more
“I use opencode consistently to build non-trivial apps with it”
Client/server model · 3
V2's client/server design and usability are praised as a clear improvement.
“I've been using OpenCode V2 since it first appeared on the website, and I can only repeat the positive views on usability”
“been using ocv2 as my daily driver for a while and had very little complaints. its a great product from a great team.”
Show 1 more
Remote and mobile · 2
With OpenChamber, its web and mobile interfaces let users work away from the desk.
“But I use OpenCode / OpenChamber for personal stuff”
Caveats and dissent
Data privacy · 3
Users report prompts sent to a cloud summarization model by default (since removed) and offline mode being hard to configure.
“opencode by default uses a "free" cloud model to summarize all chats even if a different model was configured as the main one”
“Open code is great and I use it”
Show 1 more
“OpenCode will leak all your data to themselves and to shady 3rd parties.”
Key security · 3
Commenters flag unclear API key storage and a compromised-key notice.
“there are no warnings or documentation about how they store your keys.”
“I saw the email this morning from OpenCode about my key being compromised”
Show 1 more
“I did have an account but moved off for security reasons.”
Resource usage · 3
Reports of high RAM per subagent, high CPU use and slow performance.
“it is horrible with ram usage. Every subagent takes like 300 MB of RAM.”
Show 1 more
“Performance is horrible both of the website and of the tool itself.”
Maintenance burden · 3
Some find it needs continuous debugging and dislike how issues are handled.
“I love the idea behind OpenCode but compared to managing Claude and Codex, it's such a pain and requires continuous debugging or managing.”
“What demorlised me is attitude of OpenCode devs/maintainers.”
Show 1 more
“I used V1 a lot and I am struggling with V2 currently. Yes, it just doesn't feel right.”
Context overhead · 2
A large system prompt is said to eat into the context window.
“Does it still eat all of my memory? Does it still have a super-gigantic system prompt that wastes a solid percentage of my context window”
Usage limits · 2
OpenCode Go's 5-hour rolling limit and lack of training guarantees are noted.
“its great value but the 5 hour rolling limit seems kinda annoying”
“OpenCode Go straight up doesn't give any guarantees about training on your data”
Often compared with
24 moved from Claude Code to OpenCode
2 moved from OpenCode to Claude CodePi112 comments mention both
3 moved from Pi to OpenCode
23 moved from OpenCode to Pi
3 moved from Codex to OpenCode
2 moved from OpenCode to Codex
1 moved from DeepSeek to OpenCode
6 moved from Claude to OpenCodeGLM55 comments mention both
Who recommends it in their own words
From the makers not counted in the ranking
“We support OpenCode, as well as any other coding tools accepting OpenAI compatible endpoints.”
Threads where it came up
- OpenCode 2.0 28 mentions
- Stop Using OpenCode 14 mentions
- Ask HN: What is your (AI) dev tech stack / workflow? (June 2026) 13 mentions
- OpenCode – The open source AI coding agent 13 mentions
- Ask HN: Has anyone replaced Claude/GPT with a local model for daily coding? 9 mentions
- Kimi K2.7-Code: open-source coding model with better token efficiency 9 mentions
Also written as opencode, OpenCode Go, Opencode, Opencode Go. Data: opencode.json. Quotes are verbatim; the verdict and theme labels are written by a model from these comments, and every clause links to its sources.
All comments 49
“which still uses the OpenCode backend”
“I use Claude Code, Codex, opencode and a couple other but I do not like the DX/UX.”
“I'm stripping OpenCode down to its absolute bare minimum, removing pretty much all of the vendor-specific crap”
“it runs Claude code, codex, opencode, pi, whatever model is appropriate”
“Stay away from OpenCode, its a pile of slop like Claude Code.”
“this should also apply to opencode”
“I was playing around with OpenCode on Omarchy over the weekend”
“I used V1 a lot and I am struggling with V2 currently. Yes, it just doesn't feel right.”
“I'm already comfortable using OpenCode with open models and avoid lock-in to any specific harness”
“Using OpenCode harness.”
“I use dynamic context compaction in open code plugin.”
“custom agents and subs in opencode”
“I use free models only these days. Opencode or Openrouter allows free models time to time.”
“OpenCode (for other models) under the hood”
“I use that and so far it is good for me.”
“Deepseek 4.1 Flash in OpenCode.”
“have been using it to help train opencode to drive a project for it”
“I use OpenCode at home”
“Other harnesses like OpenCode, Pi, etc. you can hook into their plugin architecture and implement your own OTEL emission”
“across models running on my sparks and OpenCode sub”
“It is still private by default”
“I did have an account but moved off for security reasons.”
“I saw the email this morning from OpenCode about my key being compromised”
“there are no warnings or documentation about how they store your keys.”
“OpenCode will leak all your data to themselves and to shady 3rd parties.”
“There are lot of other amazing open source agent harnesses like OpenCode [2] and Pi [3].”
“it is horrible with ram usage. Every subagent takes like 300 MB of RAM.”
“its great value but the 5 hour rolling limit seems kinda annoying”
“next try a harness like opencode or pi with open weight models via opencode-go for $10/m”
“via opencode-go for $10/m”
“opencode also supports more features and was build as an open alternative to claude.”
“But I use OpenCode / OpenChamber for personal stuff”
“Performance is horrible both of the website and of the tool itself.”
“I use OpenCode Go and OpenAI for almost everything”
“It uses OpenCode behind the scenes.”
“switched to deepseek-v4.1-flash from within opencode”
“what am I missing if I just use the Claude extension via vscode instead of opencode?”
“opencode server protocol which is published for the same purpose”
“I've been using OpenCode V2 since it first appeared on the website, and I can only repeat the positive views on usability”
“I used OpenCode to help me install ollama, Qwen local LLM, and connect OpenCode to it.”
“It's my daily driver - very easy to customize.”
“Since it's model agnostic, this might become my daily driver.”
“Does it still eat all of my memory? Does it still have a super-gigantic system prompt that wastes a solid percentage of my context window”
“I use only OpenCode free models and occasional 20$ sub from OpenAi or Anthropic. And get complete/working products out of them.”
“With my OpenCode Go sub of $10/month, DeepSeek is basically unlimited”
“opencode pushes CPU usage to very high”
“I use OpenCode and Pi connected to local models and to stuff like hosted DeepSeek all the time.”
“Open code is great and I use it”
“opencode by default uses a "free" cloud model to summarize all chats even if a different model was configured as the main one”