What are the hardware requirements for Power WizardLM 2 13B?

Power WizardLM 2 13B requires approximately 25 GB of VRAM and supports a context window of 32K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Power WizardLM 2 13B and how large is it?

Power WizardLM 2 13B is developed by refine-ai, 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 Power WizardLM 2 13B perform on standard benchmarks?

Power WizardLM 2 13B has the following published scores: MMLU 61.59. Compare against reference models on this page or on the LLM Explorer leaderboards.

Power WizardLM 2 13B by refine-ai — VRAM 25GB, 32K context

Name: Power WizardLM 2 13B
Author: refine-ai

Power WizardLM 2 13B is an open-source language model by refine-ai. Features: 13b LLM, VRAM: 25GB, Context: 32K, License: mit, Merged, LLM Explorer Score: 0.14, Arc: 60.5, HellaSwag: 83.8, MMLU: 61.6, GSM8K: 24.9.

Merged Model Autotrain compatible Endpoints compatible Mistral Region:us Safetensors Sharded Tensorflow Wizardlm Wizardlm2

Model Card on HF 🤗: https://huggingface.co/refine-ai/Power-WizardLM-2-13b

Power WizardLM 2 13B Parameters and Internals

Model Type

text generation

Use Cases

Areas:

text generation

Training Details

Data Sources:

dreamgen/WizardLM-2-7B used for merging multiple ranges

Methodology:

LazyMergekit used for model merging

Model Architecture:

A merge of WizardLM models with layer slicing

Input Output

Accepted Modalities:

text

LLM Name	Power WizardLM 2 13B
Repository 🤗	https://huggingface.co/refine-ai/Power-WizardLM-2-13b
Merged Model	Yes
Model Size	13b
Required VRAM	25 GB
Updated	2025-01-26
Maintainer	refine-ai
Model Type	mistral
Model Files	2.0 GB: 1-of-13 2.0 GB: 2-of-13 1.9 GB: 3-of-13 2.0 GB: 4-of-13 2.0 GB: 5-of-13 1.9 GB: 6-of-13 2.0 GB: 7-of-13 2.0 GB: 8-of-13 1.9 GB: 9-of-13 2.0 GB: 10-of-13 1.9 GB: 11-of-13 2.0 GB: 12-of-13 1.4 GB: 13-of-13
Model Architecture	MistralForCausalLM
License	mit
Context Length	32768
Model Max Length	32768
Transformers Version	4.40.1
Tokenizer Class	LlamaTokenizer
Padding Token	<unk>
Vocabulary Size	32000
Torch Data Type	float16

Rank the Power WizardLM 2 13B 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 Power WizardLM 2 13B

Best Alternatives	Context / RAM	Downloads	Likes
LuminRP 13B 128K V0.5	128K / 25.1 GB	9	2
Breeze 13B 32K Instruct V1.0	32K / 25.6 GB	8	0
Mistral V0.3 13B 32K Base V1	32K / 25.2 GB	7	0
Breeze 13B 32K Base V1.0	32K / 25.6 GB	7	0
Pandora 13B V1	32K / 24.9 GB	95	2
Pandora V1 13B	32K / 24.9 GB	5	3
BigMistral 13B GLUED	32K / 26.6 GB	15	1
BigMistral 13B	32K / 26.7 GB	1	2
...stral 13B DARE Blended FAILURE	32K / 26.7 GB	6	0
...stral 13B DARE Blended FAILURE	32K / 26.7 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than refine-ai/Power-WizardLM-2-13b.