GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
antiableofnormies
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
antiableofnormies
Model Tree
Qwen/Qwen3-8B
Input Modalities
Output Modalities
Supported Functionality
| Parameter | Value |
|---|---|
| Base model | Qwen/Qwen3-8B |
| Method | QLoRA (4-bit NF4 via Unsloth) |
| LoRA rank | r=16, alpha=16 |
| Target modules | q_proj, k_proj, v_proj, o_proj, gate_proj, up_proj, down_proj |
| Training data | 201 source files from 2 Android projects |
| File types | Kotlin (169), Gradle KTS (18), XML resources (14) |
| Epochs | 3 |
| Sequence length | 2048 |
| Learning rate | 2e-4 (cosine scheduler) |
| Hardware | T4 GPU (16GB VRAM) via Google Colab |
| Framework | Unsloth + TRL SFTTrainer |
python
from transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModel model = PeftModel.from_pretrained( AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", device_map="auto"), "antiableofnormies/qwen3-8b-lora-android-dev",)tokenizer = AutoTokenizer.from_pretrained("antiableofnormies/qwen3-8b-lora-android-dev")Usage with llama.cpppython convert_lora_to_gguf.py --base Qwen3-8B-Q4_K_M.gguf --lora ./qwen3-style-lora/ --output style-adapter.gguf llama-server -m Qwen3-8B-Q4_K_M.gguf --lora style-adapter.gguf --host 0.0.0.0 -ngl 99 --ctx-size 32768 --port 8080 --mlock # Optional: merge into a standalone GGUFllama-export-lora -m Qwen3-8B-Q4_K_M.gguf --lora style-adapter.gguf -o qwen3-code-style-merged.ggufIntended useThis adapter is designed for a local coding assistant (opencode + llama.cpp)that:- Runs entirely on your machine (no API calls)- Respects your existing code conventions- Avoids unwanted boilerplate that cloud APIs tend to add- Matches the patterns found in the training projectsLimitations- Trained on a personal codebase — style may not generalize to unrelatedprojects- LoRA rank 16 captures high-level style (naming, structure) but not deepdomain knowledge- Jetpack Compose UI code was explicitly excluded from training- English-only code and commentsfrom transformers import AutoModelForCausalLM, AutoTokenizerfrom peft import PeftModel model = PeftModel.from_pretrained( AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-8B", device_map="auto"), "antiableofnormies/qwen3-8b-lora-android-dev",)tokenizer = AutoTokenizer.from_pretrained("antiableofnormies/qwen3-8b-lora-android-dev")Usage with llama.cpppython convert_lora_to_gguf.py --base Qwen3-8B-Q4_K_M.gguf --lora ./qwen3-style-lora/ --output style-adapter.gguf llama-server -m Qwen3-8B-Q4_K_M.gguf --lora style-adapter.gguf --host 0.0.0.0 -ngl 99 --ctx-size 32768 --port 8080 --mlock # Optional: merge into a standalone GGUFllama-export-lora -m Qwen3-8B-Q4_K_M.gguf --lora style-adapter.gguf -o qwen3-code-style-merged.ggufIntended useThis adapter is designed for a local coding assistant (opencode + llama.cpp)that:- Runs entirely on your machine (no API calls)- Respects your existing code conventions- Avoids unwanted boilerplate that cloud APIs tend to add- Matches the patterns found in the training projectsLimitations- Trained on a personal codebase — style may not generalize to unrelatedprojects- LoRA rank 16 captures high-level style (naming, structure) but not deepdomain knowledge- Jetpack Compose UI code was explicitly excluded from training- English-only code and comments