From 060a47fce93f639e5fdca47ee9a22be8b5b94012 Mon Sep 17 00:00:00 2001 From: Abiba Date: Sat, 30 May 2026 13:15:19 +0000 Subject: [PATCH] revert: MoE back to 2 slots (cross-agent spread now prevents hotspot) Cross-agent GPU awareness ensures Tanko+Mumuni never simultaneously hit MoE. Second agent always overflows to Dense/VLM. MoE can safely use its extra VRAM with 2 slots since distinct agents never pile on. --- router/router.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/router/router.py b/router/router.py index b480405..242a74f 100644 --- a/router/router.py +++ b/router/router.py @@ -19,7 +19,7 @@ GPU_URLS = { } # Max concurrent requests per GPU (based on llama.cpp --parallel) GPU_MAX_CONCURRENT = { - "qwen3.6-35B-A3B": 1, # 1 slot (95C thermal emergency) + "qwen3.6-35B-A3B": 2, # 2 slots (cross-agent spread prevents overheating) "qwen3.6-27B-code": 2, # 2 slots (128K context frees VRAM) "qwen3.5-9b-vlm": 2, # 2 slots (12GB VRAM, 4GB headroom) }