GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
yinita
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
yinita
Model Tree
Qwen/Qwen3.5-9B
Input Modalities
Output Modalities
Supported Functionality
Best in-run eval was step 175 (reward -0.875, composite ~2.13), but only save_steps=50 checkpoints were kept (keep_checkpoints=1). This repo contains step 200 — the only surviving numbered checkpoint.
| Step | reward | ~composite | saved? |
|---|---|---|---|
| 175 | -0.875 | 2.13 | no (eval peak) |
| 200 | -1.009 | 1.99 | yes (this repo) |
python
from peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizerbase = "Qwen/Qwen3.5-9B"model = AutoModelForCausalLM.from_pretrained(base, torch_dtype="auto", device_map="auto")model = PeftModel.from_pretrained(model, "yinita/ps4mas-grpo-9b-sonnet-large-step200")tokenizer = AutoTokenizer.from_pretrained("yinita/ps4mas-grpo-9b-sonnet-large-step200")
For vLLM LoRA: load base Qwen3.5-9B + this adapter (see PS4MAS eval scripts).