What are the hardware requirements for Phi 2 Orange V2?

Phi 2 Orange V2 requires approximately 5.6 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Phi 2 Orange V2 and how large is it?

Phi 2 Orange V2 is developed by rhysjones, a model with 2.8b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Phi 2 Orange V2 perform on standard benchmarks?

Phi 2 Orange V2 has the following published scores: MMLU 55.72. Compare against reference models on this page or on the LLM Explorer leaderboards.

Phi 2 Orange V2 by rhysjones — VRAM 5.6GB, 2K context

Name: Phi 2 Orange V2
Author: rhysjones

Phi 2 Orange V2 is an open-source language model by rhysjones. Features: 2.8b LLM, VRAM: 5.6GB, Context: 2K, License: mit, LLM Explorer Score: 0.17, Arc: 61.9, HellaSwag: 76.3, MMLU: 55.7, GSM8K: 57.6.

Conversational Dataset:argilla/ultrafeedback-... Dataset:intel/orca dpo pairs Dataset:ldjnr/capybara Dataset:ldjnr/pure-dove Dataset:ldjnr/verified-camel Dataset:meta-math/metamathqa Dataset:migtissera/synthia-v1.... Dataset:open-orca/slimorca-ded... Endpoints compatible Model-index Phi Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/rhysjones/phi-2-orange-v2

Phi 2 Orange V2 Parameters and Internals

Model Type

text generation

Additional Notes

The model is an improved version of the original Phi-2-Orange with an updated training process and the latest updated model from Microsoft's Phi-2.

Training Details

Data Sources:

Open-Orca/SlimOrca-Dedup, migtissera/Synthia-v1.3, LDJnr/Verified-Camel, LDJnr/Pure-Dove, LDJnr/Capybara, meta-math/MetaMathQA, Intel/orca_dpo_pairs, argilla/ultrafeedback-binarized-preferences-cleaned

Input Output

Input Format:

The model uses ChatML as the prompt format, and it's recommended to always prompt with a system instruction.

Output Format:

Text format, Markdown for Python outputs.

Performance Tips:

If the model's output is overly verbose, instruct it to be short and concise.

LLM Name	Phi 2 Orange V2
Repository 🤗	https://huggingface.co/rhysjones/phi-2-orange-v2
Model Size	2.8b
Required VRAM	5.6 GB
Updated	2026-07-15
Maintainer	rhysjones
Model Type	phi
Model Files	5.0 GB: 1-of-2 0.6 GB: 2-of-2
Model Architecture	PhiForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Transformers Version	4.37.0
Tokenizer Class	CodeGenTokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	51200
Torch Data Type	float16

Rank the Phi 2 Orange V2 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Phi 2 Orange V2

Best Alternatives	Context / RAM	Downloads
MFANN3bv0.24	128K / 11.1 GB	6
MFANN3b	128K / 11.1 GB	86
MFANN3bv1.4	128K / 11.1 GB	8
MFANN Phigments Slerp V3.2	128K / 5.6 GB	6
MFANN3bv1.3	128K / 11.1 GB	14
MFANN3bv0.23	128K / 11.1 GB	10
MFANN3bv1.1	128K / 11.1 GB	10
MFANN3bv1.5	128K / 11.1 GB	8
MFANN Liminerity Slerp 4a	128K / 5.6 GB	7
MFANN3b Rebase	128K / 11.1 GB	7

Note: green Score (e.g. "73.2") means that the model is better than rhysjones/phi-2-orange-v2.