What are the hardware requirements for Orca 2 13B?

Orca 2 13B requires approximately 52.1 GB of VRAM and supports a context window of 4K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Orca 2 13B and how large is it?

Orca 2 13B is developed by microsoft, a model with 13b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Orca 2 13B perform on standard benchmarks?

Orca 2 13B has the following published scores: MMLU 60.37. Compare against reference models on this page or on the LLM Explorer leaderboards.

Orca 2 13B by microsoft — VRAM 52.1GB, 4K context

Name: Orca 2 13B
Author: microsoft

Orca 2 13B is an open-source language model by microsoft. Features: 13b LLM, VRAM: 52.1GB, Context: 4K, License: other, HF Score: 58.6, LLM Explorer Score: 0.22, Arc: 60.7, HellaSwag: 79.8, MMLU: 60.4, TruthfulQA: 56.4, WinoGrande: 76.6, GSM8K: 18.

Arxiv:2311.11045 Endpoints compatible Llama Microsoft Orca Orca2 Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/microsoft/Orca-2-13b

Orca 2 13B Benchmarks

IFEval: 31.28 vs 88 (so35)^-64.5%

ARC: 60.67 vs 96.7 (so35)^-37.3%

HellaSwag: 79.81 vs 95.3 (gpt4)^-16.3%

MMLU: 60.37 vs 88.3 (so35)^-31.6%

TruthfulQA: 56.41 vs 59 (gpt4)^-4.4%

WinoGrande: 76.64 vs 87.5 (gpt4)^-12.4%

GSM8K: 17.97 vs 96.4 (so35)^-81.4%

MATH Lvl 5: 3.17

LLME Score: 0.22193

^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)

🌟 Advertise your project 🚀

Orca 2 13B Parameters and Internals

Model Type

text generation

Use Cases

Limitations:

Inheriting limitations from LLAMA 2 and other LLMs, Data biases, Lack of contextual understanding, Lack of transparency, Content harms, Hallucination potential, Potential for misuse, Data distribution affects performance, System message induced variability, Zero-Shot setting limitations

Additional Notes

The model is not optimized for chat and lacks certain direct application capabilities without finetuning.

Training Details

Data Sources:

synthetic dataset

Methodology:

Enhanced reasoning abilities

LLM Name	Orca 2 13B
Repository 🤗	https://huggingface.co/microsoft/Orca-2-13b
Model Size	13b
Required VRAM	52.1 GB
Updated	2026-04-10
Maintainer	microsoft
Model Type	llama
Model Files	10.0 GB: 1-of-6 9.9 GB: 2-of-6 9.9 GB: 3-of-6 9.9 GB: 4-of-6 9.9 GB: 5-of-6 2.5 GB: 6-of-6
Model Architecture	LlamaForCausalLM
License	other
Context Length	4096
Model Max Length	4096
Transformers Version	4.33.1
Tokenizer Class	LlamaTokenizer
Beginning of Sentence Token	<s>
End of Sentence Token	</s>
Unk Token	<unk>
Vocabulary Size	32003
Torch Data Type	float32

Quantized Models of the Orca 2 13B

Model	Likes	Downloads	VRAM
Orca 2 13B GPTQ	27	11	7 GB
Orca 2 13B GGUF	66	909	5 GB
Orca 2 13B GGUF	7	152	5 GB
Orca 2 13B AWQ	7	11	7 GB

Best Alternatives to Orca 2 13B

Best Alternatives	Context / RAM	Downloads	Likes
Luminaura RP 13B	128K / 26 GB	7	0
Yarn Llama 2 13B 128K	128K / 26 GB	1146	112
Agent Llama2 13B 80K	80K / 26.4 GB	5	0
Chat Llama2 13B 80K	80K / 52.8 GB	5	0
LongAlign 13B 64K	64K / 26 GB	14	13
LongAlign 13B 64K	64K / 26 GB	11	13
LongAlign 13B 64K Base	64K / 26 GB	11	3
LongAlign 13B 64K Base	64K / 26 GB	6	3
Openbuddy Llama2 13B V15p1 64K	64K / 26.1 GB	6	4
Openbuddy Llama2 13b64k V15	64K / 26.1 GB	5	2

Rank the Orca 2 13B 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! 🌟

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

What open-source LLMs or SLMs are you in search of? 53205 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Check out Ag3ntum — our secure, self-hosted AI agent for server management.

Release v20260328a

Support LLM Explorer

Orca 2 13B by microsoft