MindWorkAI/AI-Studio
1
0 Fork 0
mirror of https://github.com/MindWorkAI/AI-Studio.git synced 2026-10-14 01:13:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add llmman as a self-hosted host for chat and embedding providers (#947)

Browse Source 
Co-authored-by: Thorsten Sommer <SommerEngineering@users.noreply.github.com>
This commit is contained in:
Eric CurtinandThorsten Sommer authored and GitHub committed 2026-10-13 13:52:43 +02:00
1 parent 2bd3d41421
commit 88d1f4d96a
20 files changed
+270 -89
Download Patch File Download Diff File Expand all files Collapse all files

No files matched your search

app/MindWork AI Studio/Models/Hosting/Hosts/HostSelfHosted.cs
+5 -1
View File
@@ -4,12 +4,16 @@ using AIStudio.Provider;
namespace AIStudio.Models.Hosting.Hosts;
/// <summary>
/// Somebody's own engine: Ollama, LM Studio, vLLM, llama.cpp, or a proxy in front of them.
/// Somebody's own engine: Ollama, LM Studio, vLLM, llama.cpp, llmman, or a proxy in front of them.
/// </summary>
/// <remarks>
/// vLLM serves whatever it was pointed at, and what it was pointed at is usually a hub repository:
/// "meta-llama/Llama-3.3-70B-Instruct", "01-ai/yi-large". So the organization comes off here too.
///
/// llmman goes one further and names the registry it pulled the model from:
/// "hf.co/unsloth/Qwen3.5-0.8B-GGUF:latest", "docker.io/ai/qwen3.5:0.8b". The registry is one more
/// path segment and comes off the same way, one segment at a time.
///
/// The colon does not. Ollama writes the variant after it -- "qwen3.8:27b-mlx" -- and taking that
/// off would leave a name which no longer says which build of the model is running. Only the host
/// which actually has a router treats a colon as routing.