What are the hardware requirements for Bucharest 0.2?

Bucharest 0.2 requires approximately 21.4 GB of VRAM and supports a context window of 5K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Bucharest 0.2 and how large is it?

Bucharest 0.2 is developed by Mihaiii, a model with 10.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Bucharest 0.2 perform on standard benchmarks?

Bucharest 0.2 has the following published scores: MMLU 66.03. Compare against reference models on this page or on the LLM Explorer leaderboards.

Bucharest 0.2 by Mihaiii — VRAM 21.4GB, 5K context

Name: Bucharest 0.2
Author: Mihaiii

Bucharest 0.2 is an open-source language model by Mihaiii. Features: 10.7b LLM, VRAM: 21.4GB, Context: 5K, License: apache-2.0, LLM Explorer Score: 0.12, Arc: 64.6, HellaSwag: 84.9, MMLU: 66, GSM8K: 56.8.

Base model:finetune:migtissera... Base model:migtissera/tess-10.... Conversational Dataset:mihaiii/openhermes-2.5... Llama Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/Mihaiii/Bucharest-0.2

Bucharest 0.2 Parameters and Internals

Additional Notes

The Bucharest series is mostly an experiment. Use Pallas series instead.

Training Details

Data Sources:

private dataset, Mihaiii/OpenHermes-2.5-1k-longest-curated

Methodology:

instruct based fine tune

Input Output

Input Format:

SYSTEM: USER: ASSISTANT:

LLM Name	Bucharest 0.2
Repository 🤗	https://huggingface.co/Mihaiii/Bucharest-0.2
Base Model(s)	migtissera/Tess-10.7B-v1.5b migtissera/Tess-10.7B-v1.5b
Model Size	10.7b
Required VRAM	21.4 GB
Updated	2026-07-30
Maintainer	Mihaiii
Model Type	llama
Model Files	4.9 GB: 1-of-5 5.0 GB: 2-of-5 4.9 GB: 3-of-5 4.9 GB: 4-of-5 1.7 GB: 5-of-5
Model Architecture	LlamaForCausalLM
License	apache-2.0
Context Length	6000
Model Max Length	6000
Transformers Version	4.37.0.dev0
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	bfloat16

Rank the Bucharest 0.2 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 Bucharest 0.2

Best Alternatives	Context / RAM	Downloads	Likes
SOLAR 10.7B Instruct V1.0 128K	128K / 21.4 GB	57	3
SOLAR 10.7B V1.0 Instruct 16K	16K / 21.4 GB	63	2
SOLAR 10.7B V1.0 Base 16K	16K / 21.4 GB	69	0
SauerkrautLM SOLAR Instruct	8K / 21.4 GB	62	46
PiVoT SOLAR 10.7B RP	8K / 21.4 GB	54	7
Fimbulvetr V2 Kuro Lotus Bf16	8K / 21.5 GB	6	4
MetaModelv2	8K / 21.4 GB	197	0
Kazemi 1.2 Solar	8K / 21.4 GB	0	0
Solarmer3	4K / 21.4 GB	5	0
Somer2	4K / 21.4 GB	10	0

Note: green Score (e.g. "73.2") means that the model is better than Mihaiii/Bucharest-0.2.