Downloads · 30 days
85
6% of all-time downloads
Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF
ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF
Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF is a text generation model from ermiaazarkhalili. Use it when you need the model to write or continue text. The card lists the license as gemma.
GGUF quantizations of ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth, produced via Unsloth + llama.cpp's conversion scripts.
All-time downloads
1.5K
Public
Repo size
34.6 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.gguf34.6 GB · 100%
At a glance
- Task
- Text Generation
- License
- gemma
- Access
- Public
- Created
- Apr 24, 2026
- Updated
- Apr 24, 2026
- SHA
- 234b4b4c
Try a prompt
Base models
Model card
From the Hugging Face model README
Gemma4-E4B-xLAM-Unsloth — GGUF quantized
GGUF quantizations of ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth,
produced via Unsloth + llama.cpp's conversion scripts.
| Field | Value |
|---|---|
| Source checkpoint | ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth |
| Base model | google/gemma-4-E4B-it |
| Dataset | Salesforce/xlam-function-calling-60k |
| Training | N=1 full epoch (7,500 steps, effective batch=8) |
| Conversion | Unsloth save_pretrained → llama.cpp convert_hf_to_gguf.py → llama-quantize |
| Quantization tool | llama.cpp llama-quantize (cached toolchain) |
Available quantizations
| File | Bits | Notes |
|---|---|---|
gemma4-e4b-function-calling-xlam-unsloth.q2_k.gguf | 2-bit | Smallest; aggressive quality loss |
gemma4-e4b-function-calling-xlam-unsloth.q3_k_m.gguf | 3-bit | Small; noticeable quality loss |
gemma4-e4b-function-calling-xlam-unsloth.q4_k_m.gguf | 4-bit | Recommended — best size/quality balance |
gemma4-e4b-function-calling-xlam-unsloth.q5_k_m.gguf | 5-bit | Near-full quality |
gemma4-e4b-function-calling-xlam-unsloth.q6_k.gguf | 6-bit | Very close to Q8_0 at smaller size |
gemma4-e4b-function-calling-xlam-unsloth.q8_0.gguf | 8-bit | Largest; closest to bf16 source |
Recommended default: Q4_K_M (4-bit, K-quant medium). For maximum fidelity, use Q8_0.
Usage
llama.cpp
# Text-only
llama-cli -hf ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF --jinja -p "Find flights from SFO to NYC on December 25th" -n 256
# Interactive chat
llama-cli -hf ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF --jinja -cnv
Ollama
ollama run hf.co/ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF:Q4_K_M
llama-cpp-python
from llama_cpp import Llama
llm = Llama.from_pretrained(
repo_id="ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF",
filename="*q4_k_m.gguf",
n_ctx=2048,
)
out = llm.create_chat_completion(
messages=[{"role": "user", "content": "Find flights from SFO to NYC on December 25th"}],
max_tokens=256,
)
print(out["choices"][0]["message"]["content"])
Intended use
For research and non-commercial experimentation only. Outputs should be independently verified before any downstream use.
Limitations
- GGUF quantizations have unavoidable quality loss relative to the source bfloat16 checkpoint. Use
Q5_K_M,Q6_K, orQ8_0for best fidelity. - Inherits all limitations of the source merged checkpoint (
ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth). - Limited to the 60 function schemas covered in the training dataset; performance on novel APIs may degrade. Although the base model is a vision-language model (VLM), this fine-tune targets text-only function-calling; vision capabilities are untouched but unevaluated.
Citation
@misc{ gemma4_e4b_xlam_unsloth_2026_gguf ,
author = {Ermia Azarkhalili},
title = { Gemma4-E4B-xLAM-Unsloth — GGUF quantized },
year = {2026},
publisher = {Hugging Face},
howpublished = {\url{https://huggingface.co/ermiaazarkhalili/Gemma4-E4B-Function-Calling-xLAM-Unsloth-GGUF}}
}
This gemma4 model was trained 2× faster with Unsloth and Hugging Face's TRL library.
- Task
- Text Generation
- License
- gemma
- Languages
- en
- Created
- Apr 24, 2026
- Updated
- Apr 24, 2026