GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
ewinregirgojr
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
ewinregirgojr
Model Tree
Qwen/Qwen3.8-27B
Input Modalities
Output Modalities
Supported Functionality
| Format / Quantization | Filename / Location | Size | Primary Runtime | Direct Link |
|---|---|---|---|---|
| GGUF Q8_0 | qwen3.8-9b-instruct-turbo-q8_0.gguf | 11.94 GB | Ollama, LM Studio, llama.cpp, Jan | Download GGUF |
| Safetensors Shards | model-00001 to model-00005 | 18.1 GB | PyTorch, Transformers, vLLM, TGI | Browse Files |
| LoRA Adapters | adapter_model.safetensors | 49.57 MB | PEFT, Unsloth, SFT | Adapter File |
| Checkpoint Stage | Global Steps | Training Loss | State Description | Repository Directory |
|---|---|---|---|---|
step-0 | 0 | 11.51 (Baseline) | Initial zero-shot pruned baseline | checkpoints/step-0/ |
step-50 | 50 | 4.64 | Vocabulary & syntax stabilization |
llama.cpp CLI (GGUF)bash
llama-cli -m qwen3.8-9b-instruct-turbo-q8_0.gguf \ -p "<|im_start|>user\nExplain quantum entanglement simply.<|im_end|>\n<|im_start|>assistant\n<think>\n" \ -n 256 -ngl 99Create a Modelfile:
dockerfile
FROM ./qwen3.8-9b-instruct-turbo-q8_0.ggufTEMPLATE """<|im_start|>system{{ .System }}<|im_end|><|im_start|>user{{ .Prompt }}<|im_end|><|im_start|>assistant<think>{{ .Response }}"""PARAMETER stop "<|im_start|>"PARAMETER stop "<|im_end|>"PARAMETER temperature 0.6Run:
bash
ollama create qwen3.8-9b-turbo -f ./Modelfileollama run qwen3.8-9b-turbopython
import torchfrom transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModel MODEL_ID = "ewinregirgojr/Qwen3.8-9B-Instruct-Turbo" tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)base_model = AutoModelForCausalLM.from_pretrained( MODEL_ID, torch_dtype=torch.float16, device_map="auto") # Load calibrated LoRA adaptermodel = PeftModel.from_pretrained(base_model, MODEL_ID, subfolder="checkpoints/healed-final") prompt = "<|im_start|>user\nExplain how layer pruning reduces memory bandwidth requirements during LLM inference.<|im_end|>\n<|im_start|>assistant\n<think>\n"inputs = tokenizer(prompt, return_tensors="pt").to(model.device) with torch.no_grad(): outputs = model.generate(**inputs, max_new_tokens=256, temperature=0.6, do_sample=True) print(tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True))python
from mlx_lm import load, generate model, tokenizer = load("ewinregirgojr/Qwen3.8-9B-Instruct-Turbo", adapter_path="checkpoints/healed-final")response = generate( model, tokenizer, prompt="<|im_start|>user\nWrite a fast Python function to find prime numbers using a sieve.\n<|im_end|>\n<|im_start|>assistant\n<think>\n", max_tokens=200, verbose=True)empirical_benchmark_results.json.FROM ./qwen3.8-9b-instruct-turbo-q8_0.ggufTEMPLATE """<|im_start|>system{{ .System }}<|im_end|><|im_start|>user{{ .Prompt }}<|im_end|><|im_start|>assistant<think>{{ .Response }}"""PARAMETER stop "<|im_start|>"PARAMETER stop "<|im_end|>"PARAMETER temperature 0.6checkpoints/step-50/ |
step-100 | 100 | 3.45 | Coherent sentence formation | checkpoints/step-100/ |
step-200 | 200 | 3.10 | Multi-turn reasoning alignment | checkpoints/step-200/ |
step-300 | 300 | 2.71 | Verbatim token retrieval calibration | checkpoints/step-300/ |
healed-final | 300 | 2.71 | Production-ready healed adapter weights | checkpoints/healed-final/ |