GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
groxaxo
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
groxaxo
Model Tree
groxaxo/Qwen3.5-27B-Writer-V2-Heretic-Coding-LoRA
ConicCat/Qwen3.5-27B-Writer-V2
Input Modalities
Output Modalities
Supported Functionality
Qwen3_5ForConditionalGeneration, 27B, BF16, license apache-2.0. Base commit 55f457bed78b68442d25815bd156b98f2e6fd341.0b3508764f80741f612c251c7f47b1e210ac2e63.Deliberate cross-base merge. The coding LoRA was originally trained on
llmfan46/Qwen3.5-27B-Writer-V2-uncensored-heretic. It was intentionally merged onto the cleanConicCat/Qwen3.5-27B-Writer-V2(same Writer-V2 lineage / identical architecture, vocab and module tree).
The adapter touched 607 modules. Of these, 496 language-tower modules carried trained
(non-zero) weights and were merged into the base. The remaining 111 vision-tower modules were
zero in the adapter itself (never trained) — merging them is a no-op, so the vision tower is
identical to the base model and vision behaviour should match ConicCat/Qwen3.5-27B-Writer-V2.
safe_merge=True (per-layer NaN check).B).merge_manifest.json for exact base/adapter commits and library versions.python
from transformers import AutoModelForImageTextToText, AutoProcessormodel = AutoModelForImageTextToText.from_pretrained("groxaxo/Qwevolution-V0-27B", dtype="bfloat16", device_map="auto")processor = AutoProcessor.from_pretrained("groxaxo/Qwevolution-V0-27B")