GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
aziz9788
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
aziz9788
Model Tree
unsloth/Qwen3.5-4B
Input Modalities
Output Modalities
Supported Functionality
Fully merged Qwen3.5-4B weights: unsloth/Qwen3.5-4B + aziz9788/qwen35-saudidraft-full-final-adapter @ lora_alpha=22 (direct recipe). Identity SFT has not been applied.
Load as a standalone Transformers / vLLM model — no adapter required.
Recipe: direct (Unsloth Qwen3.5-4B @ 3764fa359b9082ea5a1e4a5e3ac3aaf6e9671636 + aziz9788/qwen35-saudidraft-full-final-adapter @ lora_alpha=22, r=32, rsLoRA). Thinking off. Run: saudidraft_direct_20260822_200129. Pre-identity no-reasoning suite: /workspace/Saudi-LLM/Eval/outputs/full_suite_no_reasoning_saudidraft_direct_20260822_200129_pre_identity.