What are the hardware requirements for Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache?

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache requires approximately 27.4 GB of VRAM and supports a context window of 2K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache and how large is it?

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache is developed by saarvajanik, a model with 6.7b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache perform on standard benchmarks?

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache has the following published scores: MMLU 23.12. Compare against reference models on this page or on the LLM Explorer leaderboards.

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache by saarvajanik — VRAM 27.4GB, 2K context

Name: Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache
Author: saarvajanik

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache is an open-source language model by saarvajanik. Features: 6.7b LLM, VRAM: 27.4GB, Context: 2K, License: mit, LLM Explorer Score: 0.16, Arc: 23.8, HellaSwag: 27.1, MMLU: 23.1.

En Endpoints compatible Opt Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/saarvajanik/facebook-opt-6.7b-qcqa-ub-16-best-for-KV-cache

Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache Parameters and Internals

Model Type

text generation

Additional Notes

This version uses a modified Multi-Head Attention (MHA) architecture with 6 groups of KV heads per layer, each sharing mean-pooled K/V values for the same group instead of the original 12 separate KV heads.

Supported Languages

en (primary)

LLM Name	Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache
Repository 🤗	https://huggingface.co/saarvajanik/facebook-opt-6.7b-qcqa-ub-16-best-for-KV-cache
Model Size	6.7b
Required VRAM	27.4 GB
Updated	2026-07-22
Maintainer	saarvajanik
Model Type	opt
Model Files	10.0 GB: 1-of-3 9.9 GB: 2-of-3 7.5 GB: 3-of-3
Supported Languages	en
Model Architecture	OPTForCausalLM
License	mit
Context Length	2048
Model Max Length	2048
Transformers Version	4.23.1
Tokenizer Class	GPT2Tokenizer
Beginning of Sentence Token	</s>
End of Sentence Token	</s>
Unk Token	</s>
Vocabulary Size	50272
Torch Data Type	float32
Activation Function	relu
Errors	replace

Rank the Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache 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 Facebook Opt 6.7B Qcqa Ub 16 Best For KV Cache

Best Alternatives	Context / RAM	Downloads	Likes
...7B Gqa Ub 16 Best For KV Cache	2K / 27.4 GB	190	0
....7B Qcqa Ub 16 Best For Q Loss	2K / 27.4 GB	60	0
Galactica 6.7B ReFT GSM8k	2K / 13.8 GB	58	0
Galactica 6.7B EssayWriter	2K / 13.4 GB	55	5
Opt Flan Iml 6.7B	2K / 26.6 GB	131	1
Opt 6.7B	2K / 13.4 GB	120463	121
...lactica 6.7B Evol Instruct 70K	2K / 27.4 GB	129	19
OPT 6.7B Nerybus Mix	2K / 13.7 GB	125	20
Galactica 6.7B Finetuned	2K / 27.4 GB	89	34
OPI Galactica 6.7B	2K / 13.4 GB	116	6

Note: green Score (e.g. "73.2") means that the model is better than saarvajanik/facebook-opt-6.7b-qcqa-ub-16-best-for-KV-cache.