From 7f80a52d5ee75db5315a729cdeaa9e09040d437a Mon Sep 17 00:00:00 2001 From: Wenxi Onyx Date: Tue, 10 Mar 2026 09:51:48 -0700 Subject: [PATCH 1/4] feat: add onyx to integration page --- 4_integrations/claude-code.md | 2 +- 4_integrations/codex.md | 2 +- 4_integrations/index.md | 1 + 4_integrations/onyx.md | 125 ++++++++++++++++++++++++++++++++++ 4 files changed, 128 insertions(+), 2 deletions(-) create mode 100644 4_integrations/onyx.md diff --git a/4_integrations/claude-code.md b/4_integrations/claude-code.md index 4c63252..8c3abda 100644 --- a/4_integrations/claude-code.md +++ b/4_integrations/claude-code.md @@ -1,7 +1,7 @@ --- title: Claude Code description: Use Claude Code with LM Studio -index: 2 +index: 3 --- Claude Code can talk to LM Studio via the Anthropic-compatible `POST /v1/messages` endpoint. diff --git a/4_integrations/codex.md b/4_integrations/codex.md index 59f1258..f579e3e 100644 --- a/4_integrations/codex.md +++ b/4_integrations/codex.md @@ -1,7 +1,7 @@ --- title: Codex description: Use Codex with LM Studio -index: 3 +index: 4 --- Codex can talk to LM Studio via the OpenAI-compatible `POST /v1/responses` endpoint. diff --git a/4_integrations/index.md b/4_integrations/index.md index 8502bd8..2085424 100644 --- a/4_integrations/index.md +++ b/4_integrations/index.md @@ -13,5 +13,6 @@ We provide guides below for popular tools and are constantly expanding this list ## Available Integrations +- [Onyx](/docs/integrations/onyx) - [Claude Code](/docs/integrations/claude-code) - [Codex](/docs/integrations/codex) diff --git a/4_integrations/onyx.md b/4_integrations/onyx.md new file mode 100644 index 0000000..ce2b1cf --- /dev/null +++ b/4_integrations/onyx.md @@ -0,0 +1,125 @@ +--- +title: Onyx +description: Use Onyx with LM Studio +index: 2 +--- + +[Onyx](https://www.onyx.app/) is an open source AI platform that connects your LLMs to your organization's knowledge and applications. It provides chat, AI agents, internal search, web search, actions, and more — all in one place. + +Onyx has a built-in integration to LM Studio that auto-discovers your loaded models, so setup takes just a few clicks. + + + +```lms_protip +Have a powerful LLM rig? Use [LM Link](/docs/integrations/lmlink) to run Onyx from one machine while the model runs on another. +``` + +## Quick Start: Self-Hosted Onyx + +For full data privacy, you can self-host both Onyx and LM Studio on your own infrastructure. + +### 1) Deploy Onyx + +Run the Onyx install script (requires Bash, Docker, and at least 4 CPU cores / 8GB RAM): + +```bash +curl -fsSL https://raw.githubusercontent.com/onyx-dot-app/onyx/main/deployment/docker_compose/install.sh > install.sh && chmod +x install.sh && ./install.sh +``` + +The script will pull containers and start Onyx. Once running, access the Onyx UI at `http://localhost:3000`. + +For more deployment options (Kubernetes, cloud providers, etc.), see the [Onyx deployment docs](https://docs.onyx.app/deployment/getting_started/quickstart). + +### 2) Start LM Studio's local server + +```bash +lms server start --port 1234 +``` + +If LM Studio is running on a different machine than Onyx, make sure the LM Studio server is accessible from the Onyx host (e.g., `http://:1234`). + +### 3) Add LM Studio as a provider in Onyx + +Navigate to **Admin Panel → Configuration → LLM** and select **LM Studio** from the list of providers. + +Set the **API Base URL** to your LM Studio server address: + + +| Field | Value | +| ---------------- | ----------------------------------------------------------------------------------------- | +| **API Base URL** | `http://localhost:1234` (or `http://:1234` if on a different machine) | +| **API Key** | Leave blank unless you have [authentication](/docs/developer/core/authentication) enabled | + + +Onyx will automatically connect and discover your loaded models, including their context length and capabilities (vision, reasoning). + + + +### 4) Select your model and start chatting + +Enable your models from the auto-discovered list, save, and you're ready to go! + + +--- + +## Quick Start: Onyx Cloud + +An alternative, faster way to get started with Onyx is via [Onyx Cloud](https://cloud.onyx.app). + +Since Onyx Cloud runs remotely, it cannot access `localhost` on your machine. You'll need to expose your LM Studio server using a tunneling service like [ngrok](https://ngrok.com). + +### 1) Sign up for Onyx Cloud + +Go to [cloud.onyx.app](https://cloud.onyx.app) and sign up for a free trial — no credit card required. + +### 2) Start LM Studio's local server + +Make sure LM Studio is running as a server (default port `1234`). + +```bash +lms server start --port 1234 +``` + +```lms_protip +Use a model with more than ~25k context length. Onyx's agentic features can consume a lot of context. +``` + +### 3) Expose your LM Studio server + +Because Onyx Cloud runs on remote infrastructure, it needs a publicly reachable URL to connect to your LM Studio server. Use a tunnel to expose your local server: + +```bash +ngrok http 1234 +``` + +This will give you a public URL like `https://abcd-1234.ngrok-free.app`. Copy this URL — you'll need it in the next step. + +### 4) Add LM Studio as a provider + +In Onyx, navigate to **Admin Panel → Configuration → LLM** and select **LM Studio** from the list of providers. + +Set the **API Base URL** to the public URL from your tunnel: + +| Field | Value | +| ---------------- | ----------------------------------------------------------------------------------------- | +| **API Base URL** | Your tunnel URL (e.g., `https://abcd-1234.ngrok-free.app`) | +| **API Key** | Leave blank unless you have [authentication](/docs/developer/core/authentication) enabled | + +Onyx will automatically connect to LM Studio and discover your loaded models. + + + +### 5) Select your models and start chatting + +Enable the models you want to use in Onyx from the auto-discovered list, save, and you're ready to go! Open Onyx's chat interface and start using your local LM Studio models with agents powered by your docs. + + + +--- + +## Tips + +- Onyx auto-discovers models loaded in LM Studio along with their capabilities (vision, reasoning) and context length. +- For best results, use a model with strong instruction-following and tool-use capabilities (e.g., Qwen 3.5). + +If you're running into trouble, hop onto the [LM Studio Discord](https://discord.gg/lmstudio) or the [Onyx Discord](https://discord.gg/Pk3qzRKAEx)! From 7763b582e813bda057ad14781d4016660d6b503b Mon Sep 17 00:00:00 2001 From: Wenxi Onyx Date: Mon, 16 Mar 2026 13:11:03 -0700 Subject: [PATCH 2/4] re-organize alphabetically --- 4_integrations/claude-code.md | 2 +- 4_integrations/codex.md | 2 +- 4_integrations/index.md | 2 +- 4_integrations/onyx.md | 2 +- 4 files changed, 4 insertions(+), 4 deletions(-) diff --git a/4_integrations/claude-code.md b/4_integrations/claude-code.md index 8c3abda..5202970 100644 --- a/4_integrations/claude-code.md +++ b/4_integrations/claude-code.md @@ -1,7 +1,7 @@ --- title: Claude Code description: Use Claude Code with LM Studio -index: 3 +index: 2 --- Claude Code can talk to LM Studio via the Anthropic-compatible `POST /v1/messages` endpoint. diff --git a/4_integrations/codex.md b/4_integrations/codex.md index f579e3e..59f1258 100644 --- a/4_integrations/codex.md +++ b/4_integrations/codex.md @@ -1,7 +1,7 @@ --- title: Codex description: Use Codex with LM Studio -index: 4 +index: 3 --- Codex can talk to LM Studio via the OpenAI-compatible `POST /v1/responses` endpoint. diff --git a/4_integrations/index.md b/4_integrations/index.md index 2085424..1176710 100644 --- a/4_integrations/index.md +++ b/4_integrations/index.md @@ -13,6 +13,6 @@ We provide guides below for popular tools and are constantly expanding this list ## Available Integrations -- [Onyx](/docs/integrations/onyx) - [Claude Code](/docs/integrations/claude-code) - [Codex](/docs/integrations/codex) +- [Onyx](/docs/integrations/onyx) diff --git a/4_integrations/onyx.md b/4_integrations/onyx.md index ce2b1cf..430c915 100644 --- a/4_integrations/onyx.md +++ b/4_integrations/onyx.md @@ -1,7 +1,7 @@ --- title: Onyx description: Use Onyx with LM Studio -index: 2 +index: 4 --- [Onyx](https://www.onyx.app/) is an open source AI platform that connects your LLMs to your organization's knowledge and applications. It provides chat, AI agents, internal search, web search, actions, and more — all in one place. From 36baa72a6fc9c0366ff838ef7dc63d00b374d0d2 Mon Sep 17 00:00:00 2001 From: Wenxi Onyx Date: Wed, 25 Mar 2026 09:45:18 -0700 Subject: [PATCH 3/4] remove cloud --- 4_integrations/onyx.md | 56 ------------------------------------------ 1 file changed, 56 deletions(-) diff --git a/4_integrations/onyx.md b/4_integrations/onyx.md index 430c915..b923b96 100644 --- a/4_integrations/onyx.md +++ b/4_integrations/onyx.md @@ -59,62 +59,6 @@ Onyx will automatically connect and discover your loaded models, including their Enable your models from the auto-discovered list, save, and you're ready to go! - ---- - -## Quick Start: Onyx Cloud - -An alternative, faster way to get started with Onyx is via [Onyx Cloud](https://cloud.onyx.app). - -Since Onyx Cloud runs remotely, it cannot access `localhost` on your machine. You'll need to expose your LM Studio server using a tunneling service like [ngrok](https://ngrok.com). - -### 1) Sign up for Onyx Cloud - -Go to [cloud.onyx.app](https://cloud.onyx.app) and sign up for a free trial — no credit card required. - -### 2) Start LM Studio's local server - -Make sure LM Studio is running as a server (default port `1234`). - -```bash -lms server start --port 1234 -``` - -```lms_protip -Use a model with more than ~25k context length. Onyx's agentic features can consume a lot of context. -``` - -### 3) Expose your LM Studio server - -Because Onyx Cloud runs on remote infrastructure, it needs a publicly reachable URL to connect to your LM Studio server. Use a tunnel to expose your local server: - -```bash -ngrok http 1234 -``` - -This will give you a public URL like `https://abcd-1234.ngrok-free.app`. Copy this URL — you'll need it in the next step. - -### 4) Add LM Studio as a provider - -In Onyx, navigate to **Admin Panel → Configuration → LLM** and select **LM Studio** from the list of providers. - -Set the **API Base URL** to the public URL from your tunnel: - -| Field | Value | -| ---------------- | ----------------------------------------------------------------------------------------- | -| **API Base URL** | Your tunnel URL (e.g., `https://abcd-1234.ngrok-free.app`) | -| **API Key** | Leave blank unless you have [authentication](/docs/developer/core/authentication) enabled | - -Onyx will automatically connect to LM Studio and discover your loaded models. - - - -### 5) Select your models and start chatting - -Enable the models you want to use in Onyx from the auto-discovered list, save, and you're ready to go! Open Onyx's chat interface and start using your local LM Studio models with agents powered by your docs. - - - --- ## Tips From 0fc268fa5f41d810a16fe229714fc6461e20f8b1 Mon Sep 17 00:00:00 2001 From: Wenxi Onyx Date: Wed, 25 Mar 2026 10:05:42 -0700 Subject: [PATCH 4/4] add docs url --- 4_integrations/onyx.md | 1 + 1 file changed, 1 insertion(+) diff --git a/4_integrations/onyx.md b/4_integrations/onyx.md index b923b96..df7ec6b 100644 --- a/4_integrations/onyx.md +++ b/4_integrations/onyx.md @@ -65,5 +65,6 @@ Enable your models from the auto-discovered list, save, and you're ready to go! - Onyx auto-discovers models loaded in LM Studio along with their capabilities (vision, reasoning) and context length. - For best results, use a model with strong instruction-following and tool-use capabilities (e.g., Qwen 3.5). +- Check out the [Onyx docs](https://docs.onyx.app/) for more details on setup and configuration. If you're running into trouble, hop onto the [LM Studio Discord](https://discord.gg/lmstudio) or the [Onyx Discord](https://discord.gg/Pk3qzRKAEx)!