<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2" -->

---
title: OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!
description: A walkthrough of building a fully free AI coding agent setup using three components: OpenCode as the terminal coding agent, Nvidia NIM preview endpoints...
canonical: https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS! | daily.dev
og:description: A walkthrough of building a fully free AI coding agent setup using three components: OpenCode as the terminal coding agent, Nvidia NIM preview endpoints...
og:url: https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2
og:image: https://api.daily.dev/og/posts/VTEaWECX2.png
og:image:alt: OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!

**[AICodeKing](https://daily.dev/sources/aicodeking)** · 8 min read · 3 upvotes · 3 comments

## Summary

A walkthrough of building a fully free AI coding agent setup using three components: OpenCode as the terminal coding agent, Nvidia NIM preview endpoints (offering ~77 free models including Nemotron Ultra, MiniMax M3, and others), and Kimi K3 via ZenMLX (free for a limited time). The setup requires no paid subscriptions or API credits. The video also covers a practical model-switching strategy: Kimi K3 for long-horizon agentic tasks, Step 3.7 Flash for quick edits, MiniMax M3 for multimodal/frontend work, and Nemotron Ultra for heavy reasoning. Caveats include rate limits on Nvidia endpoints and the temporary nature of Kimi K3's free access.

## Full article

daily.dev links to this article rather than hosting it. Read it at the original source: <https://www.youtube.com/watch?v=fMzVZvK9frI>

## Questions this post answers

### How can I use Kimi K3 for free with OpenCode?

Kimi K3 can be accessed for free through ZenMLX, an AI model aggregation platform similar to OpenRouter, which lists it as Moonshot AI's Kimi-K3-Free model on a limited-time pay-as-you-go free tier. Since ZenMLX is an official provider inside OpenCode, you create a ZenMLX account, generate an API key, connect it within OpenCode, and select the free Kimi K3 model without needing to load credits.

_Anyone chasing free frontier coding models can track shifting availability of deals like this on daily.dev._

### What free models are available through Nvidia NIM preview endpoints for coding?

Nvidia's NIM catalog on build.nvidia.com hosts around 140 models with roughly 77 offered as free preview endpoints for development, including Nemotron 3 Ultra (a 550 billion parameter hybrid Mamba-transformer with a 1 million token context window), Mini Max M3 for multimodal coding and design, Step 3.7 Flash for fast edits, DeepSeek V4 Flash, GLM 5.1, and Qwen and Gemma models, all accessible via a single Nvidia API key.

_Developers piecing together no-cost model stacks can follow catalog changes like Nvidia NIM's on daily.dev._

### Why is OpenCode a good choice for connecting free AI models like Nvidia NIM and Kimi K3?

OpenCode is a free, open-source terminal coding agent that has both Nvidia and ZenMLX built in as official providers, so connecting either service only requires pasting an API key rather than editing config files or setting custom base URLs. This makes it possible to combine multiple free model sources into one model picker without any manual provider setup.

_Anyone comparing terminal coding agents can weigh setups like OpenCode's provider integrations on daily.dev._

## Community discussion

Top comments from developers on daily.dev.

**@agustinbarrientos** · 0 upvotes

> Nvidia's preview endpoints look perfect for experiments, not production.

---

Tags: [#llm](https://daily.dev/tags/llm), [#opencode](https://daily.dev/tags/opencode), [#kimi](https://daily.dev/tags/kimi)

[View this post on daily.dev](https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!","url":"https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2"},"datePublished":"2026-07-20T09:40:47.808Z","dateModified":"2026-09-14T06:26:50.074Z","description":"A walkthrough of building a fully free AI coding agent setup using three components: OpenCode as the terminal coding agent, Nvidia NIM preview endpoints...","image":"https://i.ytimg.com/vi/fMzVZvK9frI/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/fMzVZvK9frI/sddefault.jpg","isAccessibleForFree":true,"articleSection":"AICodeKing","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"AICodeKing","logo":"https://media.daily.dev/image/upload/s--x7nDUfWj--/f_auto,q_auto/v1768208312/logos/aicodeking","url":"https://daily.dev/sources/aicodeking"},"commentCount":3,"discussionUrl":"https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":3},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":3}],"keywords":"llm,opencode,kimi","timeRequired":"PT8M","video":{"@type":"VideoObject","name":"OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!","description":"A walkthrough of building a fully free AI coding agent setup using three components: OpenCode as the terminal coding agent, Nvidia NIM preview endpoints...","thumbnailUrl":"https://i.ytimg.com/vi/fMzVZvK9frI/sddefault.jpg","uploadDate":"2026-07-20T09:40:47.808Z","duration":"PT8M","url":"https://api.daily.dev/r/VTEaWECX2","embedUrl":"https://www.youtube.com/embed/fMzVZvK9frI"}}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"AICodeKing","item":"https://daily.dev/sources/aicodeking"},{"@type":"ListItem","position":3,"name":"OpenCode + FREE Kimi K3, GLM-5.2 API: IT ACTUALLY WORKS!"}]}
{"@context":"https://schema.org","@type":"WebPage","@id":"https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2","comment":[{"@type":"Comment","text":"Nvidia’s preview endpoints look perfect for experiments, not production.","datePublished":"2026-08-02T23:18:49.218Z","url":"https://daily.dev/posts/VTEaWECX2#c-toOFc4Io3","author":{"@type":"Person","name":"Agustin Barrientos","url":"https://daily.dev/agustinbarrientos","image":"https://media.daily.dev/image/upload/s--5ayxQnqn--/f_auto/v1788281802/avatars/avatar_wQYYVe5Tbj0NJ7C7qPoa8?_a=BAMAMicg0"}}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/opencode-free-kimi-k3-glm-5-2-api-it-actually-works--vteawecx2#faq","mainEntity":[{"@type":"Question","name":"How can I use Kimi K3 for free with OpenCode?","acceptedAnswer":{"@type":"Answer","text":"Kimi K3 can be accessed for free through ZenMLX, an AI model aggregation platform similar to OpenRouter, which lists it as Moonshot AI's Kimi-K3-Free model on a limited-time pay-as-you-go free tier. Since ZenMLX is an official provider inside OpenCode, you create a ZenMLX account, generate an API key, connect it within OpenCode, and select the free Kimi K3 model without needing to load credits. Anyone chasing free frontier coding models can track shifting availability of deals like this on daily.dev."}},{"@type":"Question","name":"What free models are available through Nvidia NIM preview endpoints for coding?","acceptedAnswer":{"@type":"Answer","text":"Nvidia's NIM catalog on build.nvidia.com hosts around 140 models with roughly 77 offered as free preview endpoints for development, including Nemotron 3 Ultra (a 550 billion parameter hybrid Mamba-transformer with a 1 million token context window), Mini Max M3 for multimodal coding and design, Step 3.7 Flash for fast edits, DeepSeek V4 Flash, GLM 5.1, and Qwen and Gemma models, all accessible via a single Nvidia API key. Developers piecing together no-cost model stacks can follow catalog changes like Nvidia NIM's on daily.dev."}},{"@type":"Question","name":"Why is OpenCode a good choice for connecting free AI models like Nvidia NIM and Kimi K3?","acceptedAnswer":{"@type":"Answer","text":"OpenCode is a free, open-source terminal coding agent that has both Nvidia and ZenMLX built in as official providers, so connecting either service only requires pasting an API key rather than editing config files or setting custom base URLs. This makes it possible to combine multiple free model sources into one model picker without any manual provider setup. Anyone comparing terminal coding agents can weigh setups like OpenCode's provider integrations on daily.dev."}}]}
```

