What are the hardware requirements for Meltemi 7B Instruct V1.5?

Meltemi 7B Instruct V1.5 requires approximately 15 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 Meltemi 7B Instruct V1.5 and how large is it?

Meltemi 7B Instruct V1.5 is developed by ilsp, a model with 7b 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 Meltemi 7B Instruct V1.5?

Meltemi 7B Instruct V1.5 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Meltemi 7B Instruct V1.5 by ilsp — VRAM 15GB, 32K context

Name: Meltemi 7B Instruct V1.5
Author: ilsp

Meltemi 7B Instruct V1.5 is an open-source language model by ilsp. Features: 7b LLM, VRAM: 15GB, Context: 32K, License: apache-2.0, Fine-Tuned, Instruction-Based, LLM Explorer Score: 0.14.

Arxiv:1803.05457 Arxiv:1905.07830 Arxiv:2009.03300 Arxiv:2109.07958 Arxiv:2308.16884 Arxiv:2403.07691 Arxiv:2407.20743 Conversational Deploy:azure El En Endpoints compatible Finetuned Instruct Mistral Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/ilsp/Meltemi-7B-Instruct-v1.5

Meltemi 7B Instruct V1.5 Parameters and Internals

Model Type

text generation

Supported Languages

el (high proficiency), en (high proficiency)

Training Details

Data Sources:

89,730 Greek preference data translated from high-quality datasets on Hugging Face, 7,342 English preference data

Methodology:

Odds Ratio Preference Optimization (ORPO) algorithm

Context Length:

8192

Hardware Used:

Amazon's cloud computing services via GRNET under the OCRE Cloud framework

Responsible Ai Considerations

Mitigation Strategies:

Aligned with human preferences but may generate misleading, harmful, and toxic content.

Input Output

Input Format:

Zephyr format through tokenizer's chat template

Accepted Modalities:

text

Output Format:

text

Performance Tips:

Ensure BOS token is included in tokenized prompts.

LLM Name	Meltemi 7B Instruct V1.5
Repository 🤗	https://huggingface.co/ilsp/Meltemi-7B-Instruct-v1.5
Model Size	7b
Required VRAM	15 GB
Updated	2026-08-01
Maintainer	ilsp
Model Type	mistral
Instruction-Based	Yes
Model Files	5.0 GB: 1-of-4 4.9 GB: 2-of-4 4.6 GB: 3-of-4 0.5 GB: 4-of-4
Supported Languages	el en
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.43.2
Tokenizer Class	LlamaTokenizer
Padding Token	<pad>
Vocabulary Size	61384
Torch Data Type	bfloat16

Rank the Meltemi 7B Instruct V1.5 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 Meltemi 7B Instruct V1.5

Best Alternatives	Context / RAM	Downloads	Likes
...Nemo Instruct 2407 Abliterated	1000K / 24.5 GB	220	20
SpydazWeb AI HumanAI RP	512K / 14.4 GB	16	1
SpydazWeb AI HumanAI 002	512K / 14.4 GB	18	1
...daz Web AI ChatML 512K Project	512K / 14.5 GB	12	0
... Summarize 64K QLoRANET Merged	128K / 4.1 GB	6	0
...1 Summarize 64K LoRANET Merged	128K / 14.4 GB	7	0
Mistral 7B Instruct V0.2	32K / 14.4 GB	1115337	3184
Mistral 7B Instruct V0.1	32K / 14.4 GB	253629	1846
Mistral 7B Instruct V0.2	32K / 14.4 GB	1338	0
Mixtral AI CyberCoder	32K / 14.3 GB	0	0

Note: green Score (e.g. "73.2") means that the model is better than ilsp/Meltemi-7B-Instruct-v1.5.