#Custom Model Selection

7 messages · Page 1 of 1 (latest)

quaint relic
#

Can you currently connect a custom model through the use of Base URLs. I want to use Qwen in Warp but I can’t find an option to add a custom model.

thorny trailBOT
#

Hi @quaint relic. I'm kapa.ai, your friendly support bot. I will try to answer your question by searching through the knowledge base. :wave:

#

Currently, Warp does not support connecting a custom model via a user-specified Base URL or endpoint for general users. The ability to add your own LLM (such as Qwen) by specifying a custom API endpoint is a highly requested feature, as seen in multiple GitHub issues and user discussions. Many users have asked for support to connect local models (like Qwen, Ollama, LM Studio, etc.) or to specify an OpenAI-compatible endpoint, but this is not yet available in the product for most users.

At present, Warp only allows organizations on the Enterprise plan to enable a “Bring Your Own LLM” option, where the Warp team works with you to support your preferred LLM provider. This feature is not available on other plans, and there is no UI or settings option for individual users to add a custom model or specify a Base URL for an external LLM like Qwen. The Warp team has acknowledged the demand and stated that this feature is on their roadmap, but there is no ETA for general availability yet AI FAQs, Option to add own LLM API key, Make Warp work with Local Language Models.

If you want to use Qwen or any other custom model, you will need to wait for Warp to implement this feature. For now, you can only select from the curated list of models provided by Warp (OpenAI, Anthropic, Google) in the app settings Model Choice.

Summary:
You cannot currently connect a custom model like Qwen via a Base URL in Warp. This feature is not yet available for general users, but it is a popular request and is on the roadmap.

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon:

#

@quaint relic Kapa.ai is still learning and improving, please let me know how I did by reacting below.

undone gyro
#

@thorny trail whats the github issue requesting QWEN

thorny trailBOT
#

Let me check the knowledge base for you again. One moment! :books:

#

The GitHub issue specifically requesting QWEN model support in Warp is Issue #6400: Add more QWEN model. In this issue, the user requests the addition of more LLMs, including Grok and Qwen, as they are currently not accessible in Warp. There is also a comment in the thread asking for the ability to add a custom LLM server, such as LMStudio or Ollama, but the main request is for Qwen and Grok support<source><https://github.com/warpdotdev/Warp/issues/6400</source>.>

Note: you can ask me a follow up question by @ mentioning me again :speech_balloon: