Llama3 DiscoLeo Instruct 8B 32K V0.1 By DiscoResearch: Benchmarks, Features and Detailed Analysis. Insights on Llama3 DiscoLeo Instruct 8B 32K V0.1.

Autotrain compatible Conversational De Endpoints compatible Instruct Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1

Llama3 DiscoLeo Instruct 8B 32K V0.1 Benchmarks

^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 🚀

Llama3 DiscoLeo Instruct 8B 32K V0.1 Parameters and Internals

Model Type

instruction-tuned, long-context

Use Cases

Areas:

Research, Commercial applications

Supported Languages

German (high proficiency)

Training Details

Data Sources:

ociglot-fineweb-v0.5, DiscoLM German dataset

Data Volume:

65 billion high-quality German tokens + 100 million tokens at 32k context length

Methodology:

Full fine-tuning with axolotl

Context Length:

32768

Hardware Used:

42 supercomputer at hessian.Ai

Model Architecture:

Similar to LeoLM and Occiglot models

Input Output

Input Format:

Use Llama-3 chat template with <role, content> format

Accepted Modalities:

text

Output Format:

Text output

Performance Tips:

Use appropriate GPU to leverage full potential

LLM Name	Llama3 DiscoLeo Instruct 8B 32K V0.1
Repository 🤗	https://huggingface.co/DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1
Model Size	8b
Required VRAM	16.1 GB
Updated	2025-06-09
Maintainer	DiscoResearch
Model Type	llama
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 5.0 GB: 2-of-4 4.9 GB: 3-of-4 1.2 GB: 4-of-4
Supported Languages	de
Model Architecture	LlamaForCausalLM
License	llama3
Context Length	65536
Model Max Length	65536
Transformers Version	4.40.2
Tokenizer Class	PreTrainedTokenizerFast
Padding Token	<\|end_of_text\|>
Vocabulary Size	128256
Torch Data Type	bfloat16

Llama3 DiscoLeo Instruct 8B 32K V0.1 (DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1)

Quantized Models of the Llama3 DiscoLeo Instruct 8B 32K V0.1

Model	Likes	Downloads	VRAM
... Instruct 8B 32K V0.1 4bit AWQ	0	21	5 GB

Best Alternatives to Llama3 DiscoLeo Instruct 8B 32K V0.1

Best Alternatives	Context / RAM	Downloads	Likes
...otron 8B UltraLong 4M Instruct	4192K / 32.1 GB	3284	108
UltraLong Thinking	4192K / 16.1 GB	367	2
...a 3.1 8B UltraLong 4M Instruct	4192K / 32.1 GB	176	24
...a 3.1 8B UltraLong 2M Instruct	2096K / 32.1 GB	875	9
...otron 8B UltraLong 2M Instruct	2096K / 32.1 GB	526	15
Zero Llama 3.1 8B Beta6	1048K / 16.1 GB	958	1
...otron 8B UltraLong 1M Instruct	1048K / 32.1 GB	1808	45
...a 3.1 8B UltraLong 1M Instruct	1048K / 32.1 GB	1387	29
....1 1million Ctx Dark Planet 8B	1048K / 32.3 GB	90	2
...dger Nu Llama 3.1 8B UltraLong	1048K / 16.2 GB	30	3

Note: green Score (e.g. "73.2") means that the model is better than DiscoResearch/Llama3-DiscoLeo-Instruct-8B-32k-v0.1.

Rank the Llama3 DiscoLeo Instruct 8B 32K V0.1 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? 48046 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.

Release v20241124

Support LLM Explorer