What are the hardware requirements for Jais Adapted 70B Chat?

Jais Adapted 70B Chat requires approximately 277.4 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 Jais Adapted 70B Chat and how large is it?

Jais Adapted 70B Chat is developed by inceptionai, a model with 70b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Jais Adapted 70B Chat?

Jais Adapted 70B Chat is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Jais Adapted 70B Chat by inceptionai — VRAM 277.4GB, 4K context

Name: Jais Adapted 70B Chat
Author: inceptionai

Jais Adapted 70B Chat is an open-source language model by inceptionai. Features: 70b LLM, VRAM: 277.4GB, Context: 4K, License: proprietary, LLM Explorer Score: 0.13.

Arxiv:2307.09288 Arxiv:2308.16149 Arxiv:2402.12840 Ar Arabic Base model:finetune:inceptiona... Base model:inceptionai/jais-ad... Conversational Decoder En English Jais-family Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/inceptionai/jais-adapted-70b-chat

Jais Adapted 70B Chat Parameters and Internals

Model Type

text generation, bilingual LLM, causal-lm

Use Cases

Areas:

Research, Commercial applications

Applications:

Natural language understanding, Generation tasks, Chat assistants, Sentiment analysis, Summarization

Primary Use Cases:

General AI assistant for Arabic and English, Cultural aligned language processing

Limitations:

Not suitable for multipurpose languages, Requires responsible use avoiding prohibited applications., Has limitations in handling non-target languages

Considerations:

Model should be used with understanding of its capability in Arabic and English. Not suitable for high-stakes decisions without human oversight.

Additional Notes

Extensive release focuses on Arabic NLP, model adaptation, and providing bilingual capabilities.

Supported Languages

languages_supported (Arabic, English), proficiency_level (High)

Training Details

Data Sources:

Web pages, Wikipedia articles, News articles, Social network content, Code data, Books, Scientific papers, Synthetic translation

Data Volume:

1.6 Trillion tokens

Methodology:

Pre-training from scratch and adaptive pre-training from Llama-2

Context Length:

16384

Hardware Used:

Condor Galaxy Supercomputer, 64 Cerebras CS-2 Wafer-Scale Engines

Model Architecture:

Auto-regressive transformer-based, decoder-only (GPT-3) with SwiGLU and ALiBi or RoPE and Grouped Query Attention

Safety Evaluation

Risk Categories:

Misinformation, Bias

Ethical Considerations:

Efforts to reduce bias; model may still exhibit biases and generate incorrect or misleading content.

Responsible Ai Considerations

Fairness:

Techniques employed to reduce bias

Accountability:

User is responsible for model usage and outcomes

Mitigation Strategies:

Training data curated by Inception and includes techniques to minimize bias.

Input Output

Input Format:

Text input

Accepted Modalities:

Text

Output Format:

Generates text

LLM Name	Jais Adapted 70B Chat
Repository 🤗	https://huggingface.co/inceptionai/jais-adapted-70b-chat
Base Model(s)	Jais Adapted 70B inceptionai/jais-adapted-70b
Model Size	70b
Required VRAM	277.4 GB
Updated	2024-11-04
Maintainer	inceptionai
Model Type	llama
Model Files	9.6 GB: 1-of-29 10.0 GB: 2-of-29 9.9 GB: 3-of-29 9.3 GB: 4-of-29 9.3 GB: 5-of-29 9.3 GB: 6-of-29 10.0 GB: 7-of-29 9.9 GB: 8-of-29 9.3 GB: 9-of-29 9.3 GB: 10-of-29 9.3 GB: 11-of-29 10.0 GB: 12-of-29 9.9 GB: 13-of-29 9.3 GB: 14-of-29 9.3 GB: 15-of-29 9.3 GB: 16-of-29 10.0 GB: 17-of-29 9.9 GB: 18-of-29 9.3 GB: 19-of-29 9.3 GB: 20-of-29 9.3 GB: 21-of-29 10.0 GB: 22-of-29 9.9 GB: 23-of-29 9.3 GB: 24-of-29 9.3 GB: 25-of-29 9.3 GB: 26-of-29 10.0 GB: 27-of-29 9.9 GB: 28-of-29 8.9 GB: 29-of-29
Supported Languages	ar en
Gated Model	Yes
Model Architecture	LlamaForCausalLM
License	proprietary
Context Length	4096
Model Max Length	4096
Transformers Version	4.40.1
Tokenizer Class	LlamaTokenizer
Vocabulary Size	64000
Torch Data Type	float32

Rank the Jais Adapted 70B Chat 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 Jais Adapted 70B Chat

Best Alternatives	Context / RAM	Downloads	Likes
... Chat 1048K Chinese Llama3 70B	1024K / 141.9 GB	9069	5
... Chat 1048K Chinese Llama3 70B	1024K / 141.9 GB	7684	4
... 3 70B Instruct Gradient 1048K	1024K / 141.9 GB	22	122
Llama3 Function Calling 1048K	1024K / 141.9 GB	5	1
...a 3 70B Instruct Gradient 524K	512K / 141.9 GB	24	23
...a 3 70B Instruct Gradient 262K	256K / 141.9 GB	20	56
...ama 3 70B Arimas Story RP V2.0	256K / 141.1 GB	10	3
...ama 3 70B Arimas Story RP V1.6	256K / 141.2 GB	5	0
...ama 3 70B Arimas Story RP V1.5	256K / 141.2 GB	21	3
Yi 70B 200K RPMerge Franken	195K / 142.4 GB	12	1

Note: green Score (e.g. "73.2") means that the model is better than inceptionai/jais-adapted-70b-chat.