GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
vvsotnikov
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
vvsotnikov
Model Tree
Qwen/Qwen3.8-27B
Qwen/Qwen3.6-27B
Input Modalities
Output Modalities
Supported Functionality
mtp.* drafter
tensors and vision towerQwen/Qwen3.8-27BQwen/Qwen3.6-27B revision: 6a9e13bd6fc8f0983b9b99948120bc37f49c13e9Qwen/Qwen3.8-27B revision: 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0The two source checkpoints have identical tensor names, shapes, dtypes, and
model configuration; their configs differ only in the recorded Transformers
version. The newer Qwen/Qwen3.8-27B non-weight files are used because it has the
larger merge coefficient.
Both source checkpoints are Apache-2.0. Their upstream model cards describe
capabilities, limitations, and the expected qwen3_5 runtime support.