Phi 4 8bit is an open-source language model by mlx-community. Features: 4.1b LLM, VRAM: 15.6GB, Context: 16K, License: mit, Quantized, LLM Explorer Score: 0.17.
8-bit 8bit Base model:microsoft/phi-4 Base model:quantized:microsoft... Chat Code Conversational Custom code En Endpoints compatible Math Mlx Phi Phi3 Quantized Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/mlx-community/phi-4-8bit
Phi 4 8bit Benchmarks
nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").
What is the LLM Explorer Rank (Score)
Phi 4 8bit Parameters and Internals
| LLM Name | Phi 4 8bit |
| Repository 🤗 | https://huggingface.co/mlx-community/phi-4-8bit |
| Base Model(s) | |
| Model Size | 4.1b |
| Required VRAM | 15.6 GB |
| Updated | 2025-12-31 |
| Maintainer | mlx-community |
| Model Type | phi3 |
| Model Files | |
| Supported Languages | en |
| Quantization Type | 8bit |
| Model Architecture | Phi3ForCausalLM |
| License | mit |
| Context Length | 16384 |
| Model Max Length | 16384 |
| Transformers Version | 4.47.0 |
| Tokenizer Class | GPT2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 100352 |
| Torch Data Type | bfloat16 |
Best Alternatives to Phi 4 8bit
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 4 Mini Instruct 8bit | 128K / 4.1 GB | 297 | 5 |
| Phi 3.5 Mini Instruct 4bit | 128K / 2.1 GB | 3175 | 9 |
| Numind NuExtract 1.5 MLX 8bit | 128K / 4.1 GB | 12 | 2 |
| ...hi 3 Medium 128K Instruct 4bit | 128K / 7.8 GB | 23 | 2 |
| ...m 128K Instruct 8.0bpw H8 EXL2 | 128K / 13.4 GB | 4 | 4 |
| ...m 128K Instruct 6.0bpw H6 EXL2 | 128K / 10.7 GB | 8 | 3 |
| ...dium 128K Instruct 8 0bpw EXL2 | 128K / 13.4 GB | 0 | 1 |
| ...m 128K Instruct 3.0bpw H6 EXL2 | 128K / 5.6 GB | 5 | 0 |
| ...m 128K Instruct 5.0bpw H6 EXL2 | 128K / 8.9 GB | 5 | 0 |
| Phi 3 Mini 128K Instruct 8bit | 128K / 4.1 GB | 55 | 10 |
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/phi-4-8bit .
Rank the Phi 4 8bit Capabilities
🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟
What open-source LLMs or SLMs are you in search of? 52721 in total.
Original data from HuggingFace, OpenCompass and various public git repos.
Check out Ag3ntum — our secure, self-hosted AI agent for server management.
Release v20260328a