What are the hardware requirements for Mika 7B?

Mika 7B requires approximately 14.4 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 Mika 7B and how large is it?

Mika 7B is developed by Epiculous, a model with 7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Mika 7B perform on standard benchmarks?

Mika 7B has the following published scores: MMLU 59.85. Compare against reference models on this page or on the LLM Explorer leaderboards.

Mika 7B by Epiculous — VRAM 14.4GB, 32K context

Name: Mika 7B
Author: Epiculous

Mika 7B is an open-source language model by Epiculous. Features: 7b LLM, VRAM: 14.4GB, Context: 32K, License: agpl-3.0, Instruction-Based, LLM Explorer Score: 0.13, Arc: 63.5, HellaSwag: 85.4, MMLU: 59.9, GSM8K: 28.5.

Conversational Dataset:chaoticneutrals/synthe... Dataset:chaoticneutrals/synthe... Dataset:epiculous/gnosis Dataset:grimulkan/theory-of-mi... Dataset:lemonilia/limarp Endpoints compatible Instruct Mistral Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/Epiculous/Mika-7B

Mika 7B Parameters and Internals

Additional Notes

Mika is named after a Claude-3 Opus chat persona.

Training Details

Data Sources:

lemonilia/LimaRP, grimulkan/theory-of-mind, Epiculous/Gnosis, ChaoticNeutrals/Synthetic-RP, ChaoticNeutrals/Synthetic-Dark-RP

Methodology:

Trained in a similar manner to Fett-uccine with synthetic RP data created by Claude included.

Input Output

Input Format:

ChatML Context Template, Mistral Instruct Template

Performance Tips:

YMMV (Your Mileage May Vary)

LLM Name	Mika 7B
Repository 🤗	https://huggingface.co/Epiculous/Mika-7B
Model Size	7b
Required VRAM	14.4 GB
Updated	2026-07-31
Maintainer	Epiculous
Model Type	mistral
Instruction-Based	Yes
Model Files	4.9 GB: 1-of-3 5.0 GB: 2-of-3 4.5 GB: 3-of-3
Model Architecture	MistralForCausalLM
License	agpl-3.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.38.2
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Mika 7B 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 Mika 7B

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	5	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 Epiculous/Mika-7B.