From ab13f2c19fcf0c299bc32923dcf6d781b31c7bbf Mon Sep 17 00:00:00 2001 From: "russell@unturf.com" Date: Tue, 27 Jan 2026 08:12:07 -0500 Subject: [PATCH] Disable gpt-oss.ai.unturf.com endpoint (returns 404) --- public/src/config.js | 11 +---------- 1 file changed, 1 insertion(+), 10 deletions(-) diff --git a/public/src/config.js b/public/src/config.js index bb4d2bc..d4890a4 100644 --- a/public/src/config.js +++ b/public/src/config.js @@ -178,16 +178,7 @@ export const VLLM_ENDPOINTS = [ 'hf.co/unsloth/Qwen3-Coder-30B-A3B-Instruct-GGUF:Q4_K_M': 32768, } }, - { - id: "gpt-oss.ai.unturf.com", - url: "https://gpt-oss.ai.unturf.com/v1", - // Ollama endpoints don't return max_model_len, so we specify it here - // Note: Model supports 262144 but limited by GPU VRAM (24GB) - modelContextWindows: { - 'gpt-oss:latest': 64000, - 'deepseek-r1:14b-qwen-distill-q8_0': 64000, - } - }, + // { id: "gpt-oss.ai.unturf.com", url: "https://gpt-oss.ai.unturf.com/v1" }, // Disabled - upstream returns 404 ]; // Helper to get vault value or localStorage fallback