What are the hardware requirements for BigWeave V12 90B?

BigWeave V12 90B requires approximately 175.7 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 BigWeave V12 90B and how large is it?

BigWeave V12 90B is developed by llmixer, a model with 90b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

How does BigWeave V12 90B perform on standard benchmarks?

BigWeave V12 90B has the following published scores: MMLU 69.41. Compare against reference models on this page or on the LLM Explorer leaderboards.

BigWeave V12 90B by llmixer — VRAM 175.7GB, 4K context

Name: BigWeave V12 90B
Author: llmixer

BigWeave V12 90B is an open-source language model by llmixer. Features: 90b LLM, VRAM: 175.7GB, Context: 4K, License: llama2, LLM Explorer Score: 0.12, Arc: 68.1, HellaSwag: 87.7, MMLU: 69.4, GSM8K: 47.4.

90b Conversational Dare En Endpoints compatible Euryale 1.3 Frankenmerge Llama Model-index Platypus2 Region:us Safetensors Sharded Tensorflow Ties Wintergoddess Xwin

Model Card on HF 🤗: https://huggingface.co/llmixer/BigWeave-v12-90b

BigWeave V12 90B Parameters and Internals

Model Type

text generation

Additional Notes

The BigWeave models aim to identify merge settings equaling or surpassing the performance of Goliath-120b. The merged model achieves lower perplexity than each component model separately. The 90b size allows for 4bit quants to fit into 48GB of VRAM.

LLM Name	BigWeave V12 90B
Repository 🤗	https://huggingface.co/llmixer/BigWeave-v12-90b
Model Size	90b
Required VRAM	175.7 GB
Updated	2026-07-24
Maintainer	llmixer
Model Type	llama
Model Files	9.8 GB: 1-of-18 9.7 GB: 2-of-18 9.9 GB: 3-of-18 9.7 GB: 4-of-18 9.8 GB: 5-of-18 9.8 GB: 6-of-18 10.0 GB: 7-of-18 9.8 GB: 8-of-18 9.8 GB: 9-of-18 9.8 GB: 10-of-18 10.0 GB: 11-of-18 9.6 GB: 12-of-18 9.7 GB: 13-of-18 9.8 GB: 14-of-18 9.8 GB: 15-of-18 10.0 GB: 16-of-18 9.6 GB: 17-of-18 9.1 GB: 18-of-18
Supported Languages	en
Model Architecture	LlamaForCausalLM
License	llama2
Context Length	4096
Model Max Length	4096
Transformers Version	4.36.2
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32000
Torch Data Type	float16

Rank the BigWeave V12 90B 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 BigWeave V12 90B

Best Alternatives	Context / RAM	Downloads	Likes
Chytrej 90M Base	8K / 0.2 GB	302	2
Chytrej1.5 90M Base	8K / 0.2 GB	11	2
Chytrej1 90M Base	8K / 0.2 GB	16	2
BigWeave V6 90B	4K / 175.7 GB	83	0
BigWeave V9 90B	4K / 175.7 GB	4	1
Tinyllama Zh	2K / 4.9 GB	8	1

Note: green Score (e.g. "73.2") means that the model is better than llmixer/BigWeave-v12-90b.