Phi 1 5 Base is an open-source language model by voidful. Features: 1.5b LLM, VRAM: 6.2GB, Context: 2K, License: mit, LLM Explorer Score: 0.15, Arc: 50.3, HellaSwag: 61.2, MMLU: 39.3, GSM8K: 16.5.
Arxiv:1910.09700 Autotrain compatible Custom code Endpoints compatible Phi Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/voidful/phi-1_5_base
Phi 1 5 Base Benchmarks
ARC: 50.26 vs 96.7 (so35)-48%
HellaSwag: 61.2 vs 95.3 (gpt4)-35.8%
MMLU: 39.34 vs 88.3 (so35)-55.4%
GSM8K: 16.53 vs 96.4 (so35)-82.9%
LLME Score: 0.15322
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 1 5 Base Parameters and Internals
| Model Type |
| |||
| Additional Notes |
| |||
| Training Details |
|
| LLM Name | Phi 1 5 Base |
| Repository 🤗 | https://huggingface.co/voidful/phi-1_5_base |
| Model Size | 1.5b |
| Required VRAM | 6.2 GB |
| Updated | 2025-08-18 |
| Maintainer | voidful |
| Model Type | phi |
| Model Files | |
| Model Architecture | PhiForCausalLM |
| License | mit |
| Context Length | 2048 |
| Model Max Length | 2048 |
| Transformers Version | 4.39.1 |
| Tokenizer Class | CodeGenTokenizer |
| Vocabulary Size | 80980 |
| Torch Data Type | float32 |
Best Alternatives to Phi 1 5 Base
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 1 5 Chat 128k | 2K / 3.1 GB | 660 | 6 |
| ...etuned Medicine Conversational | 0K / 2.8 GB | 14 | 1 |
Rank the Phi 1 5 Base 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? 54964 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