GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
jsantillana
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
jsantillana
Model Tree
Qwen/Qwen2.5-7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
| Model | Params | B1 KW | B2 F1 | B3 TM | B4 Tool | B5 Chat |
|---|---|---|---|---|---|---|
| VectraYX-Nano v7 (headline) | 42M | 0.332±0.005 | — | — | 0.230±0.052 | 0.725±0.130 |
| VectraYX-Pro 3B | 3.2B | 0.341 | 0.695 | 0.686 | 0.600 | 0.800 |
| VectraYX-Pro 7B | 7B | 0.335 | 0.815 | 0.686 | 0.880 | 0.800 |
| GPT-4o (frontier ref.) | — | 0.333 | 0.110 | 0.520 | 0.615 | 0.631 |
B4=0.880 — best tool-selection score in the VectraYX family. B2=0.815 (best threat classification).
This adapter applies the VectraYX cybersecurity specialization to Qwen2.5-7B-Instruct:
ml.g5.xlarge)<|tool_call|> emission (B4=0.880, highest in family)python
from peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch # Load 4-bit quantized base model (~5 GB VRAM)bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_quant_type="nf4")base_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen2.5-7B-Instruct", quantization_config=bnb_config, device_map="auto") # Load VectraYX QLoRA adapter on topmodel = PeftModel.from_pretrained(base_model, "jsantillana/vectrayx-pro-7b")tokenizer = AutoTokenizer.from_pretrained("jsantillana/vectrayx-pro-7b") # Inferencemessages = [{"role": "user", "content": "¿Cuáles son los CVEs más críticos relacionados con Log4j?"}]text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=300, temperature=0.7)print(tokenizer.decode(outputs[0], skip_special_tokens=True))| Model | Backbone | Params | B4 Tool | B2 F1 |
|---|---|---|---|---|
| VectraYX-Nano v7 | from-scratch | 42M | 0.230±0.052 | — |
| VectraYX-Base | from-scratch | 260M | 0.000* | 0.220 |
| VectraYX-Pro 3B | Qwen2.5-3B + LoRA-64 | 3.2B | 0.600 | 0.695 |
| VectraYX-Pro 7B | Qwen2.5-7B + QLoRA-32 | 7B | 0.880 | 0.815 |
bibtex
@misc{santillana2026vectrayx, title = {VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity Language Model with Curriculum Learning and Native Tool Use}, author = {Santillana, Juan S.}, year = {2026}, eprint = {2605.13989}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2605.13989}}from peft import PeftModelfrom transformers import AutoModelForCausalLM, AutoTokenizer, BitsAndBytesConfigimport torch # Load 4-bit quantized base model (~5 GB VRAM)bnb_config = BitsAndBytesConfig( load_in_4bit=True, bnb_4bit_compute_dtype=torch.bfloat16, bnb_4bit_quant_type="nf4")base_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen2.5-7B-Instruct", quantization_config=bnb_config, device_map="auto") # Load VectraYX QLoRA adapter on topmodel = PeftModel.from_pretrained(base_model, "jsantillana/vectrayx-pro-7b")tokenizer = AutoTokenizer.from_pretrained("jsantillana/vectrayx-pro-7b") # Inferencemessages = [{"role": "user", "content": "¿Cuáles son los CVEs más críticos relacionados con Log4j?"}]text = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)inputs = tokenizer(text, return_tensors="pt").to(model.device)outputs = model.generate(**inputs, max_new_tokens=300, temperature=0.7)print(tokenizer.decode(outputs[0], skip_special_tokens=True))@misc{santillana2026vectrayx, title = {VectraYX-Nano: A 42M-Parameter Spanish Cybersecurity Language Model with Curriculum Learning and Native Tool Use}, author = {Santillana, Juan S.}, year = {2026}, eprint = {2605.13989}, archivePrefix = {arXiv}, primaryClass = {cs.CL}, url = {https://arxiv.org/abs/2605.13989}}