Skip to content

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4.1, MiniMax-H3, Gemma 4, FLUX and more.

Studio: update external llama.cpp default port to 9931 (#14266)

oobabooga56d6f7563a34
diff --git a/studio/backend/core/inference/providers.py b/studio/backend/core/inference/providers.py
index dd4ef5902ad..8118239207e 100644
--- a/studio/backend/core/inference/providers.py
+++ b/studio/backend/core/inference/providers.py
@@ -457,7 +457,7 @@ PROVIDER_REGISTRY: dict[str, dict[str, Any]] = {
},
"llama_cpp": {
"display_name": "llama.cpp",
- "base_url": "http://localhost:8080/v1",
+ "base_url": "http://localhost:9931/v1",
"default_models": [],
"supports_streaming": True,
"supports_vision": True,
diff --git a/studio/frontend/src/features/chat/external-providers.ts b/studio/frontend/src/features/chat/external-providers.ts
index 7cdcb9e65dd..94e9bfab387 100644
--- a/studio/frontend/src/features/chat/external-providers.ts
+++ b/studio/frontend/src/features/chat/external-providers.ts
@@ -417,7 +417,7 @@ export const CUSTOM_PROVIDER_PRESETS = [
{
providerType: "llama_cpp",
displayName: "llama.cpp",
- baseUrlPlaceholder: "http://localhost:8080/v1",
+ baseUrlPlaceholder: "http://localhost:9931/v1",
modelIdsPlaceholder: "gpt-oss-20b\nqwen3-14b",
},
{