Similar Models

unsloth/Phi-4-mini-reasoning-GGUF

Microsoft Phi-4-mini distilled for step-by-step reasoning. 3.8B params, 128K context, MIT license. Unsloth bug-fixed GGUF for reliable agentic tool-calling.

reasoningagentictool-calling

mistralai/Ministral-3-3B-Instruct-GGUF

New

Mistral AI edge-optimized 3.4B+0.4B vision model. Native function calling, JSON outputs, 256K context. Built for tool-using agentic pipelines.

agent-readyvision-textfunction-calling

ggml-org/SmolLM3-3B-GGUF

New

A 2026-native 3B reasoning model from Hugging Face. Dual-mode `/think` and `/no_think` for agentic workflows with 64K-128K context. Fully open recipe.

reasoningagenticdual-mode

← Back to Models

Phi-4-mini-instruct (3.8B Reasoning)

unsloth/Phi-4-mini-instruct-GGUF

Microsoft Phi-4-mini distilled for edge reasoning. 3.8B params, 128K context, MIT license. Optimized for agentic tool-calling and multilingual tasks.

How to Use

To get started, install the `transformers` library:

pip install transformers

Then, use the following snippet to load the model:

from transformers import AutoTokenizer, AutoModelForCausalLM

model_id = "unsloth/Phi-4-mini-instruct-GGUF"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id)

# Your inference code here...

Available Versions

Tag / Variant	Size	Format	Download
unsloth/Phi-4-mini-instruct-GGUF:Q4_K_M	2.8GB	GGUF	Link
unsloth/Phi-4-mini-instruct-GGUF:Q5_K_S	3.2GB	GGUF	Link

Model Details

Teacher Model

Phi-4-14B

Distillation Method

Knowledge Distillation (Logits)

Training Dataset

Flickr30k (Conceptual)

Primary Task

Multimodal Generation

Performance Metrics (Example)

Metric	Student Model	Teacher Model
Model Size	2.8GB	8.5GB
BLEU Score	28.5	30.1