What are the hardware requirements for MistralLite 11B?

MistralLite 11B requires approximately 21.4 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 MistralLite 11B and how large is it?

MistralLite 11B is developed by NurtureAI, a model with 11b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does MistralLite 11B perform on standard benchmarks?

MistralLite 11B has the following published scores: MMLU 50.09. Compare against reference models on this page or on the LLM Explorer leaderboards.

MistralLite 11B by NurtureAI — VRAM 21.4GB, 32K context

Name: MistralLite 11B
Author: NurtureAI

MistralLite 11B is an open-source language model by NurtureAI. Features: 11b LLM, VRAM: 21.4GB, Context: 32K, License: apache-2.0, LLM Explorer Score: 0.14, Arc: 57.7, HellaSwag: 79.5, MMLU: 50.1, GSM8K: 0.4.

Autotrain compatible Mistral Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/NurtureAI/MistralLite-11B

MistralLite 11B Parameters and Internals

Model Type

text-generation

Use Cases

Areas:

long context line and topic retrieval, summarization, question-answering

Limitations:

Requires custom container for input context > 12K tokens

Additional Notes

Supports long context (up to 32K tokens) with improved rotary embedding and sliding window techniques.

Supported Languages

English (native)

Training Details

Data Sources:

SLidingEncoder and Decoder (SLED), (Long) Natural Questions (NQ), OpenAssistant Conversations Dataset (OASST1)

Context Length:

32768

Input Output

Input Format:

<|prompter|>Your prompt here~~<|assistant|>

Accepted Modalities:

text

Output Format:

generated text

Performance Tips:

Use a pre-allocated Docker container for >12K tokens.

LLM Name	MistralLite 11B
Repository 🤗	https://huggingface.co/NurtureAI/MistralLite-11B
Model Size	11b
Required VRAM	21.4 GB
Updated	2024-12-08
Maintainer	NurtureAI
Model Type	mistral
Model Files	4.9 GB: 1-of-5 5.0 GB: 2-of-5 4.9 GB: 3-of-5 4.9 GB: 4-of-5 1.7 GB: 5-of-5
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Padding Token	[PAD]
Vocabulary Size	32003
Torch Data Type	float16

Rank the MistralLite 11B 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 MistralLite 11B

Best Alternatives	Context / RAM	Downloads	Likes
Starling LM 11B Alpha	32K / 21.4 GB	100	15
...elik V2.3 Instruct MedIT Merge	32K / 22.3 GB	1592	2
Normistral 11B Translate	32K / 22.8 GB	193	1
ConfigurableBeagle 11B	32K / 21.4 GB	8206	3
CarbonBeagle 11B	32K / 21.4 GB	8331	9
...elik V2.3 Instruct Llama Prune	32K / 15.4 GB	1387	0
CarbonBeagle 11B Truthy	32K / 21.4 GB	7713	10
Mistral 11B Instruct V0.1	32K / 42.9 GB	169	2
Mistral Evolved 11B V0.1	32K / 22.3 GB	58	33
Mistral 11B Miniplatypus	32K / 21.5 GB	9	0

Note: green Score (e.g. "73.2") means that the model is better than NurtureAI/MistralLite-11B.