What are the hardware requirements for UNA SOLAR 10.7B Instruct V1.0?

UNA SOLAR 10.7B Instruct V1.0 requires approximately 21.4 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 UNA SOLAR 10.7B Instruct V1.0 and how large is it?

UNA SOLAR 10.7B Instruct V1.0 is developed by fblgit, 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 UNA SOLAR 10.7B Instruct V1.0 perform on standard benchmarks?

UNA SOLAR 10.7B Instruct V1.0 has the following published scores: MMLU 66.1. Compare against reference models on this page or on the LLM Explorer leaderboards.

UNA SOLAR 10.7B Instruct V1.0 by fblgit — VRAM 21.4GB, 4K context

Name: UNA SOLAR 10.7B Instruct V1.0
Author: fblgit

UNA SOLAR 10.7B Instruct V1.0 is an open-source language model by fblgit. Features: 10.7b LLM, VRAM: 21.4GB, Context: 4K, License: cc-by-nc-nd-4.0, Quantized, Instruction-Based, LLM Explorer Score: 0.11, Arc: 70.7, HellaSwag: 88.3, MMLU: 66.1, GSM8K: 63.4.

Alignment-handbook Base model:quantized:upstage/s... Base model:upstage/solar-10.7b... Conversational Doi:10.57967/hf/1514 En Endpoints compatible Generated from trainer Ggml Gguf Instruct Llama Q5 Quantized Region:us Safetensors Sharded Single-turn Tensorflow Una

Model Card on HF 🤗: https://huggingface.co/fblgit/UNA-SOLAR-10.7B-Instruct-v1.0

UNA SOLAR 10.7B Instruct V1.0 Parameters and Internals

Model Type

single-turn

Additional Notes

This model, UNA-SOLAR-10.7B-Instruct-v1.0, is an alignment-oriented SFT (Supervised Fine-Tuning) model that builds on the SOLAR base model by incorporating unique tuning layers, allowing for improved regularization and customization.

Supported Languages

English (Proficient)

Training Details

Model Architecture:

SFT Further with Linear adjustments and Fan in/out modifications

LLM Name	UNA SOLAR 10.7B Instruct V1.0
Repository 🤗	https://huggingface.co/fblgit/UNA-SOLAR-10.7B-Instruct-v1.0
Base Model(s)	upstage/SOLAR-10.7B-Instruct-v1.0 upstage/SOLAR-10.7B-Instruct-v1.0
Model Size	10.7b
Required VRAM	21.4 GB
Updated	2026-07-09
Maintainer	fblgit
Model Type	llama
Instruction-Based	Yes
Model Files	7.6 GB 8.8 GB 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
Supported Languages	en
GGML Quantization	Yes
GGUF Quantization	Yes
Quantization Type	ggml\|q5\|gguf\|q5_k
Model Architecture	LlamaForCausalLM
License	cc-by-nc-nd-4.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.35.2
Tokenizer Class	LlamaTokenizer
Padding Token	</s>
Vocabulary Size	32000
Torch Data Type	float16

Rank the UNA SOLAR 10.7B Instruct V1.0 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 UNA SOLAR 10.7B Instruct V1.0

Best Alternatives	Context / RAM	Downloads	Likes
...ndle SOLAR 10.7B Instruct V1.0	4K / 21.4 GB	54	0
SOLAR 10.7B Instruct V1.0 GGUF	4K / 4.5 GB	581	2
... 10.7B Instruct V1.0 128K EXL2	128K / 5.6 GB	5	1
Eclectus 1.1 Dedup	4K / 21.4 GB	5	0
Eclectus1.1	4K / 21.4 GB	6	0
Solar 10.7B SLERP EXL2	4K / 5.6 GB	11	4
...Instruct V1.0 6.0bpw H6 EXL2 2	4K / 8.2 GB	5	2
...Instruct V1.0 3.0bpw H6 EXL2 2	4K / 4.3 GB	6	0
...Instruct V1.0 8.0bpw H8 EXL2 2	4K / 10.8 GB	3	1
SOLAR 10.7B Instruct V1.0 128K	128K / 21.4 GB	57	3

Note: green Score (e.g. "73.2") means that the model is better than fblgit/UNA-SOLAR-10.7B-Instruct-v1.0.