GLM-5.2 is live. #1 throughput on OpenRouter, pay-per-token on FriendliAI. Try it today ➜
RYVR
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Run this model inference with full control and performance in your environment.
Model Details
Model Provider
RYVR
Model Tree
Qwen/Qwen2.5-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModelbase_id = "Qwen/Qwen2.5-7B-Instruct"adapter_id = "RYVR/qwen2.5-7b-b2b-marketing-lora"tokenizer = AutoTokenizer.from_pretrained(adapter_id)model = AutoModelForCausalLM.from_pretrained(base_id, torch_dtype="auto", device_map="auto")model = PeftModel.from_pretrained(model, adapter_id)messages = [{"role": "system", "content": "You are an expert B2B marketing copywriter."},{"role": "user", "content": "Write a 3-email nurture sequence for a mid-market SaaS platform selling revenue-cycle automation to hospital CFOs."},]inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, return_tensors="pt").to(model.device)outputs = model.generate(inputs, max_new_tokens=1024, temperature=0.7)print(tokenizer.decode(outputs[0][inputs.shape[1]:], skip_special_tokens=True))
bash
vllm serve Qwen/Qwen2.5-7B-Instruct \--enable-lora \--lora-modules b2b-marketing=RYVR/qwen2.5-7b-b2b-marketing-lora
r=16, alpha=32, dropout=0.05q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_projThe adapter inherits the capabilities and biases of Qwen2.5-7B-Instruct and reflects the stylistic conventions of its B2B marketing training corpus (predominantly English, western business idiom). Generated marketing claims are not fact-checked; human review is required before any generated content is published.
Released under Apache 2.0, consistent with the Qwen2.5-7B-Instruct base model.