What are the hardware requirements for Coma 7B V0.1?

Coma 7B V0.1 requires approximately 13.6 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 Coma 7B V0.1 and how large is it?

Coma 7B V0.1 is developed by DevQuasar, 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 Coma 7B V0.1?

Coma 7B V0.1 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Coma 7B V0.1 by DevQuasar — VRAM 13.6GB, 4K context

Name: Coma 7B V0.1
Author: DevQuasar

Coma 7B V0.1 is an open-source language model by DevQuasar. Features: 7b LLM, VRAM: 13.6GB, Context: 4K, License: llama2, Code Generating, LLM Explorer Score: 0.13.

Arxiv:2203.05482 Autotrain compatible Base model:merge:meta-llama/co... Base model:merge:meta-llama/ll... Base model:meta-llama/codellam... Base model:meta-llama/llama-2-... Codegen Endpoints compatible Llama Merge Mergekit Model-index Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/DevQuasar/coma-7B-v0.1

Coma 7B V0.1 Parameters and Internals

Additional Notes

Experiment to try merged models using mergekit.

Training Details

Methodology:

Merge of pre-trained models using linear merge method

Model Architecture:

Merge of meta-llama/Llama-2-7b-hf and meta-llama/CodeLlama-7b-hf models

LLM Name	Coma 7B V0.1
Repository 🤗	https://huggingface.co/DevQuasar/coma-7B-v0.1
Base Model(s)	Llama 2 7B Hf CodeLlama 7B Hf meta-llama/Llama-2-7b-hf meta-llama/CodeLlama-7b-hf
Model Size	7b
Required VRAM	13.6 GB
Updated	2025-11-13
Maintainer	DevQuasar
Model Type	llama
Model Files	0.9 GB: 1-of-14 1.0 GB: 2-of-14 1.0 GB: 3-of-14 1.0 GB: 4-of-14 1.0 GB: 5-of-14 0.9 GB: 6-of-14 1.0 GB: 7-of-14 1.0 GB: 8-of-14 1.0 GB: 9-of-14 1.0 GB: 10-of-14 0.9 GB: 11-of-14 1.0 GB: 12-of-14 1.0 GB: 13-of-14 0.9 GB: 14-of-14
Generates Code	Yes
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	4096
Model Max Length	4096
Transformers Version	4.41.2
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	float16

Rank the Coma 7B V0.1 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 Coma 7B V0.1

Best Alternatives	Context / RAM	Downloads	Likes
Deepseek Coder 6.7B Instruct	16K / 13.5 GB	535311	507
CodeLlama 7B Hf	16K / 13.5 GB	254150	377
CodeLlama 7B Instruct Hf	16K / 13.5 GB	58587	258
GetCode Slerp	16K / 13.6 GB	87	1
Wizardllama 7B	16K / 13.5 GB	48	0
GoLLIE 7B	16K / 40.5 GB	2779	32
Codellama 7B Hf ReFT GSM8k	16K / 13.5 GB	459	1
MathCoder2 CodeLlama 7B	16K / 13.5 GB	10	5
Stack Codellama 7B Inst	16K / 13.5 GB	66	0
CodeLlama 7B Python Hf	16K / 13.5 GB	1935	146

Note: green Score (e.g. "73.2") means that the model is better than DevQuasar/coma-7B-v0.1.