| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
Discuss (mail)
|
| Community Apps Dependencies | Show |
| Lines of code | 21442 |
| Technical Name |
muk_ai_compat |
| License | See License tab |
| Website | http://www.mukit.at |
| Availability |
Odoo Online
Odoo.sh
On Premise
|
| Odoo Apps Dependencies |
Discuss (mail)
|
| Community Apps Dependencies | Show |
| Lines of code | 21442 |
| Technical Name |
muk_ai_compat |
| License | See License tab |
| Website | http://www.mukit.at |
MuK AI Compatible Providers
Ollama, vLLM, OpenRouter & every OpenAI-compatible LLM
MuK IT GmbH - www.mukit.at
Run any AI you like, right from MuK AI
Adds OpenAI-compatible LLMs to MuK AI. Run private models on your own hardware with Ollama, LM Studio, vLLM or llama.cpp, or tap into a cloud aggregator like OpenRouter, Groq, Together AI, Fireworks AI or LiteLLM. The chat, agents, tools and approval flow of MuK AI all work unchanged — just point them at the model of your choice.
Adding a new endpoint takes 30 seconds: pick OpenAI Compatible, paste the URL, drop in the API key, hit Sync Models, done. No Python, no module release, no waiting on us.
Bring your own model
Use a model you already host. Use a model your team prefers. Mix and match: an office Ollama for everyday work, a cloud aggregator for the heavy lifting, a dedicated vLLM box for batch jobs. Every endpoint that speaks the OpenAI Chat dialect plugs straight into MuK AI — tool calls, multimodal attachments, streamed answers and usage tracking all behave the same.
An Ollama record ships pre-configured
for http://localhost:11434 — install
Ollama, activate the record, click Sync, pick a model and
chat. Everything else is added from the Providers form
in under a minute.
Tested endpoints
Copy the URL, pick the auth scheme, save. Tested out of the box:
| Endpoint | Code | API URL | Auth | Manager |
|---|---|---|---|---|
| Ollama | ollama |
http://localhost:11434/v1 |
None | Ollama |
| LM Studio | lm_studio |
http://localhost:1234/v1 |
None | None |
| vLLM | vllm |
http://localhost:8000/v1 |
Bearer | None |
| llama.cpp | llama_cpp |
http://localhost:8080/v1 |
None | None |
| OpenRouter | openrouter |
https://openrouter.ai/api/v1 |
Bearer | None |
| Groq | groq |
https://api.groq.com/openai/v1 |
Bearer | None |
| Together AI | together |
https://api.together.xyz/v1 |
Bearer | None |
| Fireworks AI | fireworks |
https://api.fireworks.ai/inference/v1 |
Bearer | None |
| LiteLLM Proxy | litellm_proxy |
http://localhost:4000/v1 |
Bearer | None |
OpenRouter asks for a HTTP-Referer and an
X-Title header? Drop them into the
Extra Headers field as a JSON object — any
future endpoint that wants a custom header works the same
way.
Self-hosted with Ollama
Run everything inside your network. Pulled llama3.3, qwen2.5 or deepseek-r1 on your Ollama box? Click Sync Models — the local catalog populates instantly and the models show up in the agent picker.
Pulling new models or removing old ones is a single
command on the Ollama host
(ollama pull llama4:70b,
ollama rm llama3.3). Sync again afterward
to refresh the catalog. No giant downloads tunneled
through the Odoo UI, no waiting on a browser tab.
Clean reasoning for open-source models
Open-source reasoning models — DeepSeek-R1, QwQ, Qwen 3 Reasoning and their distills — think out loud before answering. Without help, that internal monologue lands right in the chat next to the answer and clutters every conversation.
The addon catches it and tucks it away in the same reasoning panel MuK AI uses for GPT-5 and Claude — the answer stays clean, the thinking is one click away when you want to inspect it.
Help and Support
Feel free to contact us, if you need any help with your Odoo
integration or additional features.
You will get 30 days of
support in case of any issues (except data recovery, migration or
training).
Our Services
Odoo
Development
Odoo
Integration
Odoo
Infrastructure
Odoo
Training
Odoo
Support
MuK AI Compatible Providers
Extends muk_ai with support for any OpenAI-compatible LLM endpoint — self-hosted servers (Ollama, LM Studio, vLLM, llama.cpp) and cloud aggregators (OpenRouter, Groq, Together AI, Fireworks AI, LiteLLM proxy). Ollama records use the native GET /api/tags for Sync Models; pull and remove stay in the Ollama CLI.
One translator class powers every endpoint; everything else is records. New endpoints can be configured from the Providers UI without writing code or shipping a release.
Installation
To install this module, you need to:
Download the module and add it to your Odoo addons folder. Afterward, log on to your Odoo server and go to the Apps menu. Trigger the debug mode and update the list by clicking on the "Update Apps List" link. Now install the module by clicking on the install button.
Upgrade
To upgrade this module, you need to:
Download the module and add it to your Odoo addons folder. Restart the server and log on to your Odoo server. Select the Apps menu and upgrade the module by clicking on the upgrade button.
What ships
A single Selection value — OpenAI Compatible — covers every endpoint that speaks POST /v1/chat/completions.
A single pre-shipped record — Ollama — points at the standard http://localhost:11434/v1, no authentication, and has the Ollama manager enabled so Sync Models hits /api/tags. The record is archived by default; activate it once Ollama is running on the host.
Any other endpoint is a 30-second UI add.
Adding endpoints from the UI
Navigate to MuK AI > Configuration > Providers > Create. Pick OpenAI Compatible and fill in the API URL, auth scheme, and any extra headers required by the upstream service.
| Endpoint | Code | API URL | Auth |
|---|---|---|---|
| LM Studio | lm_studio | http://localhost:1234/v1 | None |
| vLLM | vllm | http://localhost:8000/v1 | Bearer |
| llama.cpp | llama_cpp | http://localhost:8080/v1 | None |
| OpenRouter | openrouter | https://openrouter.ai/api/v1 | Bearer |
| Groq | groq | https://api.groq.com/openai/v1 | Bearer |
| Together AI | together | https://api.together.xyz/v1 | Bearer |
| Fireworks AI | fireworks | https://api.fireworks.ai/inference/v1 | Bearer |
| LiteLLM Proxy | litellm_proxy | http://localhost:4000/v1 | Bearer |
OpenRouter wants HTTP-Referer and X-Title headers — drop those in the Extra Headers field as a JSON object ({"HTTP-Referer": "https://your.host", "X-Title": "Odoo"}).
Click Sync Models — the catalog populates from GET /v1/models.
Ollama Sync
Records with Manager = Ollama swap the model-list endpoint from GET /v1/models to Ollama's native GET /api/tags, which is the only path that returns the locally-pulled tags. Sync Models on the provider upserts local muk_ai.model records for each one.
Pulling new models or removing old ones is done from the Ollama CLI (ollama pull llama4:70b, ollama rm llama3.3). Re-run Sync Models afterward to refresh the catalog. The chat API works regardless of manager_kind — you can point a Manager = None record at an Ollama server's /v1 endpoint and use it purely as a chat provider.
Reasoning models (<think> blocks)
OSS reasoning models — DeepSeek-R1, QwQ, Qwen-3-Reasoning, their distills — emit chain-of-thought wrapped in <think>...</think> inside the regular content stream. Without handling, the block bleeds into the assistant message, the conversation history balloons, and the chat UI shows the model's internal monologue as text.
The provider record has Surface <think> Reasoning enabled by default. When set, the addon strips <think>...</think> from the visible content and routes it as a reasoning event to the existing muk_ai reasoning panel — same surface that OpenAI's reasoning.encrypted_content lands on. No-op for models that do not emit <think> tags, so it is safe to leave on.
Credits
Contributors
- Mathias Markl <mathias.markl@mukit.at>
Author & Maintainer
This module is maintained by the MuK IT GmbH.
MuK IT is an Austrian company specialized in customizing and extending Odoo. We develop custom solutions for your individual needs to help you focus on your strength and expertise to grow your business.
If you want to get in touch please contact us via mail (sale@mukit.at) or visit our website (https://mukit.at).
MuK Proprietary License v1.0 This software and associated files (the "Software") may only be used (executed, modified, executed after modifications) if you have purchased a valid license from MuK IT GmbH. The above permissions are granted for a single database per purchased license. Furthermore, with a valid license it is permitted to use the software on other databases as long as the usage is limited to a testing or development environment. You may develop modules based on the Software or that use the Software as a library (typically by depending on it, importing it and using its resources), but without copying any source code or material from the Software. You may distribute those modules under the license of your choice, provided that this license is compatible with the terms of the MuK Proprietary License (For example: LGPL, MIT, or proprietary licenses similar to this one). It is forbidden to publish, distribute, sublicense, or sell copies of the Software or modified copies of the Software. The above copyright notice and this permission notice must be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
Please log in to comment on this module