Phi 3 Mini 4K Instruct V0.3 is an open-source language model by MaziyarPanahi. Features: 3.8b LLM, VRAM: 7.7GB, Context: 4K, License: mit, Instruction-Based, HF Score: 70.3, LLM Explorer Score: 0.22, Arc: 63.5, HellaSwag: 80.9, MMLU: 69.2, TruthfulQA: 60.7, WinoGrande: 72.8, GSM8K: 74.5.
Autotrain compatible Axolotl Base model:microsoft/phi-3-min... Chatml Code Conversational Custom code Dpo En Finetuned Instruct License:mit Microsoft Model-index Phi Phi-3 Phi3 Pytorch Region:us Safetensors Sharded Tensorflow
Model Card on HF 🤗: https://huggingface.co/MaziyarPanahi/Phi-3-mini-4k-instruct-v0.3
Phi 3 Mini 4K Instruct V0.3 Benchmarks
MMLU Pro: 31.42
GPQA: 9.06
MUSR: 7.75
BBH: 37.66
IFEval: 49.26 vs 88 (so35)-44%
ARC: 63.48 vs 96.7 (so35)-34.4%
HellaSwag: 80.86 vs 95.3 (gpt4)-15.2%
MMLU: 69.24 vs 88.3 (so35)-21.6%
TruthfulQA: 60.66 vs 59 (gpt4)2.8%
WinoGrande: 72.77 vs 87.5 (gpt4)-16.8%
GSM8K: 74.53 vs 96.4 (so35)-22.7%
MATH Lvl 5: 2.95
LLME Score: 0.21592
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 3 Mini 4K Instruct V0.3 Parameters and Internals
| LLM Name | Phi 3 Mini 4K Instruct V0.3 |
| Repository 🤗 | https://huggingface.co/MaziyarPanahi/Phi-3-mini-4k-instruct-v0.3 |
| Model Creator | MaziyarPanahi |
| Base Model(s) | |
| Model Size | 3.8b |
| Required VRAM | 7.7 GB |
| Updated | 2024-07-20 |
| Maintainer | MaziyarPanahi |
| Model Type | phi3 |
| Instruction-Based | Yes |
| Model Files | |
| Supported Languages | en |
| Model Architecture | Phi3ForCausalLM |
| License | mit |
| Context Length | 4096 |
| Model Max Length | 4096 |
| Transformers Version | 4.40.0.dev0 |
| Tokenizer Class | LlamaTokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 32064 |
| Torch Data Type | bfloat16 |
Quantized Models of the Phi 3 Mini 4K Instruct V0.3
Model |
Likes |
Downloads |
VRAM |
|---|---|---|---|
| ...i 3 Mini 4K Instruct V0.3 GGUF | 5 | 227 | 1 GB |
| ...Instruct V0.3 HQQ 1bit Smashed | 0 | 5 | 0 GB |
Best Alternatives to Phi 3 Mini 4K Instruct V0.3
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 3.5 Mini Instruct | 128K / 7.7 GB | 291298 | 909 |
| MediPhi Instruct | 128K / 7.7 GB | 4024 | 65 |
| NuExtract V1.5 | 128K / 7.7 GB | 108511 | 89 |
| Phi 3 Mini 128K Instruct | 128K / 7.7 GB | 5 | 0 |
| Phi 3.5 Mini TitanFusion 0.1 | 128K / 7.7 GB | 5 | 0 |
| Phi 4 Mini Instruct | 128K / 7.7 GB | 8381 | 24 |
| ECE EIFFEL 3Bv2 | 128K / 7.7 GB | 5 | 0 |
| NuExtract 1.5 | 128K / 7.7 GB | 4283 | 245 |
| MediPhi Clinical | 128K / 7.7 GB | 385 | 13 |
| MediPhi MedCode | 128K / 7.7 GB | 208 | 7 |
Note: green Score (e.g. "73.2") means that the model is better than MaziyarPanahi/Phi-3-mini-4k-instruct-v0.3 .
Rank the Phi 3 Mini 4K Instruct V0.3 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