What are the hardware requirements for Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0?

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 requires approximately 0.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 and how large is it?

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 is developed by ToBeStyled, a model with 27b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0?

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 by ToBeStyled — VRAM 0.9GB

Name: Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0
Author: ToBeStyled

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 is an open-source language model by ToBeStyled. Features: 27b LLM, VRAM: 0.9GB, License: apache-2.0, Quantized, LLM Explorer Score: 0.28.

5090 Agent Base model:agentionai/signal-3... Base model:quantized:agentiona... Benchmark Blackwell Coding Conversational Dflash Dflash2 Endpoints compatible Gguf Gsm8k Image-text-to-text Imatrix Llama.cpp Multilingual Multimodal Nvfp4 Q4 Quantized Qwen Qwen3 Qwen3.8 Qwen3 5 Reasoning Region:us Rtx-5090 Speculative-decoding Vision

Model Card on HF 🤗: https://huggingface.co/ToBeStyled/Signal-3.8-27B-NVFP4-Blackwell-DFlash2-Ultra-V1.0

Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0 Parameters and Internals

LLM Name	Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0
Repository 🤗	https://huggingface.co/ToBeStyled/Signal-3.8-27B-NVFP4-Blackwell-DFlash2-Ultra-V1.0
Base Model(s)	agentionai/Signal-3.8-27B agentionai/Signal-3.8-27B
Model Size	27b
Required VRAM	0.9 GB
Updated	2026-09-15
Maintainer	ToBeStyled
Model Type	qwen3_5
Model Files	16.9 GB 1.1 GB 0.9 GB
GGUF Quantization	Yes
Quantization Type	gguf\|q4\|q4_k
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Transformers Version	5.8.0.dev0

Rank the Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra 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 Signal 3.8 27B NVFP4 Blackwell DFlash2 Ultra V1.0

Best Alternatives	Context / RAM	Downloads	Likes
...N V1.1 NM DAU NEO MAX MTP GGUF	0K / 0.9 GB	421538	328
Qwen3.8	0K / 0.9 GB	54	1
... Opus Reasoning Distilled GGUF	0K / 0.9 GB	67921	679
Qwen3.8 27B GGUF	0K / 0.9 GB	0	685
Qwen3.8 27B OBLITERATED	0K / 83.2 GB	2402	2
...B Abliterated V2 PQ2 0 MTP MLX	0K / 0.8 GB	2349	7
...N Blackwell DFlash2 Ultra V1.0	0K / 0.9 GB	3531	3
Swift Qwen3.8 27B MIX GGUF	0K / 0.9 GB	2219	2
...inkingCap Qwen3.8 27B MIX GGUF	0K / 13.4 GB	1395	7
Qwen3.8 27B Growpath	0K / 54.7 GB	351	0

Note: green Score (e.g. "73.2") means that the model is better than ToBeStyled/Signal-3.8-27B-NVFP4-Blackwell-DFlash2-Ultra-V1.0.