Fine-tuned version of google/gemma-4-26B-A4B-it for reliable tool use and function calling.
Training
- Base model: google/gemma-4-26B-A4B-it (Mixture of Experts)
- Fine-tuning framework: Unsloth
- Hardware: NVIDIA A100 80GB (HuggingFace Space)
- Method: QLoRA (4-bit) → merged to 16-bit
Training Data
- NousResearch/hermes-function-calling-v1 — 1,893 examples of
structured tool use and function calling in Hermes format - teknium/OpenHermes-2.5 — 5,000 sampled examples for general instruction following and reasoning
Total: 6,893 examples, 2 epochs
Training Results
| Step | Loss |
|---|---|
| 10 | 1.825 |
| 50 | 0.374 |
| 200 | 0.196 |
| 500 | 0.110 |
| 862 | 0.113 |
Final training loss: 0.224
Intended Use
Designed for agentic pipelines requiring reliable structured tool call generation. Tested with Ollama for local inference.
Files
model-0000x-of-00002.safetensors— merged 16-bit weightsgemma4-hermes-tools-Q4_K_M.gguf— quantized for local inference via Ollama/llama.cpp
License
Inherits Gemma Terms of Use