GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
TheOneWhoWill
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
TheOneWhoWill
Model Tree
google/gemma-4-12B-it
Input Modalities
Output Modalities
Supported Functionality
For the training of this model I used the QLoRA method with a 16GB 5070 Ti. It 3 hours and 48 minutes to complete (480 steps) however previous protyping totaled 30 hours aggregate. The training loss went from 1.695 to 0.120 and the held-out eval loss was 0.206. I could've trained for much longer but I ran into overfitting issues on previous attempts so I decided that training on a small concentrated dataset was the best option. The training dataset consisted of 5,211 high-quality agentic/tool-calling examples, every one formatted with Gemma 4's own chat template. The glaiveai was the bulk of the data but I also included AgentInstruct examples because they had real world bash and terminal actions. I converted those raw ReAct trajectories (bash, SQL, web, and knowledge-graph actions) into real structured Gemma tool calls instead of plain text. I split it 98/2 into 5,107 train / 104 held-out val, and the eval set was built only from the val split so none of the 102 scored prompts were ever seen during training preventing contamination.
The breakdown by source:
| Source | Examples |
|---|---|
glaiveai/glaive-function-calling-v2 | 3,347 |
zai-org/AgentInstruct — alfworld | 336 |
zai-org/AgentInstruct — webshop | 351 |
zai-org/AgentInstruct — os | 195 |
zai-org/AgentInstruct — mind2web | 120 |
zai-org/AgentInstruct — db | 538 |
zai-org/AgentInstruct — kg | 324 |
| Total | 5,211 |
That's ~5,161 actual tool-calling examples (~99% of the set) trained over 3 epochs with QLoRA r=16/alpha=32, LR 2e-4 cosine, and an effective batch size of 32.
This model has fp16, Q8_0, Q6_K, Q5_K_M, and Q4_K_M quantization available in the gguf format. You can find them in this repository
To use this model with llama.cpp you can use the ggufs like so
bash
llama-server \-m gemma-coder-Q4_K_M.gguf-c 32768
Ollama requires you to create a model file with each gguf so you could have something like this
markdown
FROM ./gemma-coder-Q4_K_M.ggufPARAMETER num_ctx 32768
then you can run it like so with the Modelfile in the same directory as the gguf
bash
ollama create coding-monkey -f Modelfile
Deploy this model with Lorivo for a production-ready, OpenAI-compatible API endpoint. Lorivo supports serving fine-tuned LoRA adapters without requiring you to merge the adapter into the base model or manually manage the serving infrastructure. Note: Lorivo currently supports the LoRA adapter format containing adapter_config.json and the associated .safetensors adapter weights. The merged model or GGUF files are not supported for this deployment method.
bash
lorivo deploy ./Coding-Monkey-Gemma