GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
XReyRobert
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
XReyRobert
Model Tree
Jackrong/Qwopus3.6-27B-v2
Input Modalities
Output Modalities
Supported Functionality
XReyRobert/Qwopus3.6-27B-v2-GPTQ-Pro-v1Jackrong/Qwopus3.6-27B-v2mtp.* tensors: 15{"bfloat16": 15}18.22 GiB total safetensorsThis is meant for loader and vLLM speculative-decoding experiments. Previous testing on 1x RTX 3090 showed that restoring MTP made draft acceptance work, but did not improve throughput versus the non-MTP GPTQ-Pro baseline. The likely bottleneck was vLLM/GPTQ-Marlin speculative path overhead rather than MTP tensor precision.
For practical long-context 1x RTX 3090 serving, the non-MTP baseline remains the recommended artifact:
text
XReyRobert/Qwopus3.6-27B-v2-GPTQ-Pro-v1
Structural checks performed during patch creation:
model.safetensors.index.json contains restored mtp.* keys.config.json records mtp_num_hidden_layers >= 1.Runtime serving validation is still required before treating this as a working MTP deployment artifact.