Puffin Phi V2 is an open-source language model by teknium. Features: LLM, VRAM: 2.8GB, License: other, LLM Explorer Score: 0.1.
Model Card on HF 🤗: https://huggingface.co/teknium/Puffin-Phi-v2
Puffin Phi V2 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)
Puffin Phi V2 Parameters and Internals
| Model Type |
| |||||||||
| Additional Notes |
| |||||||||
| Training Details |
| |||||||||
| Input Output |
|
| LLM Name | Puffin Phi V2 |
| Repository 🤗 | https://huggingface.co/teknium/Puffin-Phi-v2 |
| Required VRAM | 2.8 GB |
| Updated | 2026-03-13 |
| Maintainer | teknium |
| Model Type | mixformer-sequential |
| Model Files | |
| Supported Languages | en |
| Model Architecture | MixFormerSequentialForCausalLM |
| License | other |
| Model Max Length | 2048 |
| Transformers Version | 4.34.0.dev0 |
| Tokenizer Class | CodeGenTokenizer |
| Vocabulary Size | 50304 |
| Torch Data Type | bfloat16 |
| Activation Function | gelu_new |
Best Alternatives to Puffin Phi V2
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Phi 2 | 0K / 11.1 GB | 16 | 22 |
| The OverThinker Phi 1 5 | 0K / 2.8 GB | 7 | 7 |
| Phi Ko1ep | 0K / 2.8 GB | 5 | 0 |
| Phasmid 1 5 V0.5 | 0K / 2.8 GB | 19 | 0 |
| Scarlett Phi | 0K / 2.8 GB | 6 | 8 |
| IF PromptMKR Phi | 0K / 2.8 GB | 3 | 2 |
| Oo Phi 1 5 | 0K / 2.8 GB | 41 | 32 |
| Samantha Phi | 0K / 2.8 GB | 11 | 27 |
| Samantha Phi | 0K / 2.8 GB | 8 | 27 |
| Phi Gath | 0K / 2.8 GB | 6 | 0 |
Note: green Score (e.g. "73.2") means that the model is better than teknium/Puffin-Phi-v2 .
Rank the Puffin Phi V2 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? 52473 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