GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
AMaslovskyi
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
AMaslovskyi
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
Qwen/Qwen3-8Bpython
from transformers import AutoTokenizer, AutoModelForCausalLMfrom peft import PeftModel # Load base modelbase_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3-8B", torch_dtype="auto", device_map="auto")tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B") # Load LoRA adaptermodel = PeftModel.from_pretrained(base_model, "AMaslovskyi/qwen-devops-foundation-lora") # Use the modelprompt = "How do I deploy a Kubernetes cluster?"inputs = tokenizer(prompt, return_tensors="pt")outputs = model.generate(**inputs, max_length=200, temperature=0.7)response = tokenizer.decode(outputs[0], skip_special_tokens=True)print(response)| Category | Score | Rating | Comments |
|---|---|---|---|
| CI/CD Pipelines | 1.00 | 🏆 Perfect | Complete GitHub Actions mastery, build automation |
| Docker Security | 0.75 | ✅ Strong | Production security practices, container optimization |
| Troubleshooting | 0.75 | ✅ Strong | Systematic debugging, log analysis, event investigation |
| Kubernetes Deployment | 0.25 | ❌ Needs Work | Limited deployment strategies, service configuration |
| Infrastructure as Code | 0.25 | ❌ Needs Work | Basic IaC concepts, needs more Terraform/Ansible |
| Metric | Fine-tuned Model | Base Qwen3-8B | Improvement |
|---|---|---|---|
| Response Time | 40.4s | 55.1s | 🏆 +26% Faster |
| DevOps Relevance | 6.0/10 | 6.8/10 | ⚠️ Specialized focus |
| Specialization | High | General | ✅ DevOps-focused |
| Platform | Status | Performance | Notes |
|---|---|---|---|
| Apple Silicon (M1/M2/M3) | ✅ Excellent | Fast inference | CPU-optimized, MPS supported |
| Intel/AMD x86-64 | ✅ Excellent | Good performance | 16+ cores recommended |
| NVIDIA GPU | ✅ Optimal | Fastest inference | RTX 4090/5090, A100, H100 |
| AMD GPU | ⚠️ Limited | Basic support | ROCm required, experimental |
| Device Type | RAM | Performance | Use Case |
|---|---|---|---|
| High-end Laptop | 32-64GB | 🟢 Excellent | Development, personal use |
| Workstation | 64GB+ | 🟢 Optimal | Team deployment, production |
| Cloud Instance | 32GB+ | 🟢 Scalable | API serving, multiple users |
| Entry Laptop | 16-24GB | 🟡 Limited | Light testing only |
bash
# Core requirementstorch>=2.0.0transformers>=4.35.0peft>=0.5.0 # Optional but recommendedaccelerate>=0.24.0bitsandbytes>=0.41.0 # For quantizationflash-attn>=2.0.0 # For GPU optimization🥇 Excellent Performance:
✅ Strong Performance:
🎯 Ideal For:
This model was trained on DevOps-related datasets including:
Perfect for personal use or small teams with sufficient hardware:
python
import torchfrom transformers import AutoTokenizer, AutoModelForCausalLMfrom peft import PeftModel # Optimized for local deploymentbase_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3-8B", torch_dtype=torch.float16, device_map="cpu", # Use "auto" if you have GPU trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")model = PeftModel.from_pretrained(base_model, "AMaslovskyi/qwen-devops-foundation-lora") # DevOps-optimized generationdef ask_devops_expert(question): prompt = f"<|im_start|>system\nYou are a DevOps expert. Provide practical, actionable advice.<|im_end|>\n<|im_start|>user\n{question}<|im_end|>\n<|im_start|>assistant\n" inputs = tokenizer(prompt, return_tensors="pt") outputs = model.generate( **inputs, max_length=512, temperature=0.7, do_sample=True, pad_token_id=tokenizer.eos_token_id ) response = tokenizer.decode(outputs[0], skip_special_tokens=True) return response[len(prompt):].strip() # Example usageprint(ask_devops_expert("How do I set up a CI/CD pipeline with GitHub Actions?"))Docker Container:
dockerfile
FROM python:3.11-slimRUN pip install torch transformers peft# Copy your inference scriptCMD ["python", "inference_server.py"]API Server:
✅ Ready For:
⚠️ Monitor For:
adapter_model.safetensors: LoRA adapter weights (main model file)adapter_config.json: LoRA configuration parameterstokenizer.json: Fast tokenizer configurationtokenizer_config.json: Tokenizer settings and parametersspecial_tokens_map.json: Special token mappingsvocab.json: Vocabulary mappingmerges.txt: BPE merge rulesApache 2.0
This model has been comprehensively evaluated across 21 DevOps scenarios with:
Evaluation Tools Available:
CI/CD Pipeline Setup:
text
User: How do I set up a CI/CD pipeline with GitHub Actions?Model: I'll help you set up a complete CI/CD pipeline with GitHub Actions...[Provides step-by-step workflow configuration, testing stages, deployment automation]Docker Security:
text
User: What are Docker security best practices for production?Model: Here are the essential Docker security practices for production environments...[Covers non-root users, image scanning, minimal base images, secrets management]Troubleshooting:
text
User: My Kubernetes pod is stuck in Pending state. How do I troubleshoot?Model: Let's systematically troubleshoot your pod scheduling issue...[Provides kubectl commands, event analysis, resource checking steps]from transformers import AutoTokenizer, AutoModelForCausalLMfrom peft import PeftModel # Load base modelbase_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3-8B", torch_dtype="auto", device_map="auto")tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B") # Load LoRA adaptermodel = PeftModel.from_pretrained(base_model, "AMaslovskyi/qwen-devops-foundation-lora") # Use the modelprompt = "How do I deploy a Kubernetes cluster?"inputs = tokenizer(prompt, return_tensors="pt")outputs = model.generate(**inputs, max_length=200, temperature=0.7)response = tokenizer.decode(outputs[0], skip_special_tokens=True)print(response)# Core requirementstorch>=2.0.0transformers>=4.35.0peft>=0.5.0 # Optional but recommendedaccelerate>=0.24.0bitsandbytes>=0.41.0 # For quantizationflash-attn>=2.0.0 # For GPU optimizationimport torchfrom transformers import AutoTokenizer, AutoModelForCausalLMfrom peft import PeftModel # Optimized for local deploymentbase_model = AutoModelForCausalLM.from_pretrained( "Qwen/Qwen3-8B", torch_dtype=torch.float16, device_map="cpu", # Use "auto" if you have GPU trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-8B")model = PeftModel.from_pretrained(base_model, "AMaslovskyi/qwen-devops-foundation-lora") # DevOps-optimized generationdef ask_devops_expert(question): prompt = f"<|im_start|>system\nYou are a DevOps expert. Provide practical, actionable advice.<|im_end|>\n<|im_start|>user\n{question}<|im_end|>\n<|im_start|>assistant\n" inputs = tokenizer(prompt, return_tensors="pt") outputs = model.generate( **inputs, max_length=512, temperature=0.7, do_sample=True, pad_token_id=tokenizer.eos_token_id ) response = tokenizer.decode(outputs[0], skip_special_tokens=True) return response[len(prompt):].strip() # Example usageprint(ask_devops_expert("How do I set up a CI/CD pipeline with GitHub Actions?"))FROM python:3.11-slimRUN pip install torch transformers peft# Copy your inference scriptCMD ["python", "inference_server.py"]User: How do I set up a CI/CD pipeline with GitHub Actions?Model: I'll help you set up a complete CI/CD pipeline with GitHub Actions...[Provides step-by-step workflow configuration, testing stages, deployment automation]User: What are Docker security best practices for production?Model: Here are the essential Docker security practices for production environments...[Covers non-root users, image scanning, minimal base images, secrets management]User: My Kubernetes pod is stuck in Pending state. How do I troubleshoot?Model: Let's systematically troubleshoot your pod scheduling issue...[Provides kubectl commands, event analysis, resource checking steps]