GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
CloudGoat
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
CloudGoat
Model Tree
BAAI/AREX-Turbo
Jackrong/Qwopus3.5-4B-v3
InternScience/Agents-A1-4B
Jackrong/Qwopus3.5-4B-Coder
Input Modalities
Output Modalities
Supported Functionality
I performed the merge according to these YAML files:]
yaml
slices:- sources:- model: Jackrong/Qwopus3.5-4B-v3layer_range: [0, 32]- model: Jackrong/Qwopus3.5-4B-Coderlayer_range: [0, 32]merge_method: slerpbase_model: Jackrong/Qwopus3.5-4B-v3parameters:t:- filter: self_attnvalue: [0.2, 0.3, 0.4, 0.5, 0.6, 0.6, 0.5]- filter: mlpvalue: [0.5, 0.5, 0.5, 0.5, 0.5]- value: 0.5dtype: bfloat16
yaml
slices:- sources:- model: BAAI/AREX-Turbolayer_range: [0, 32]weight: 0.5- model: InternScience/Agents-A1-4Blayer_range: [0, 32]weight: 0.5merge_method: slerpbase_model: BAAI/AREX-Turboparameters:t:- value: 0.5dtype: bfloat16
yaml
merge_method: dare_tiesbase_model: /path/to/Qwen/Qwen3.5-4Bmodels:- model: /home/CloudGoat/llms_merge/Qwents-4Bparameters:weight: 1.0density: 1.0- model: /home/CloudGoat/llms_merge/Qwopus3.5-4B-Fusionparameters:weight: 0.7density: 0.3parameters:int8_mask: truenormalize: falsedtype: bfloat16
In progress