ST

Step 3.5 Flash MTP

StepGGUF by tnhnyzcUpdated 2026-06-03
Download on Hugging Face
Parameters1.87B
Context256K
Layers46
Architecturestep35
Smallest quant2.0 GB
Quick answer

Step 3.5 Flash MTP needs about 91.2 GB of memory at IQ3_XXS (a 81.0 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 4 quants, from 2.0 GB to 96.3 GB (IQ4_XS).

Step 3.5 Flash MTP is a 1.87B-parameter Step model in GGUF format, quantized by tnhnyzc. Its trained context length is 256K tokens. The repository has 4 quant files, from 2.0 GB to 96.3 GB (IQ4_XS).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—2.0 GB 2.7 GB
Fits Get file →
IQ3_XXS81.0 GB (2 parts) 91.2 GB
Too big Get file →
IQ3_S89.9 GB (3 parts) 101.2 GB
Too big Get file →
IQ4_XS96.3 GB (3 parts) 108.4 GB
Too big Get file →

What GPU do you need for Step 3.5 Flash MTP?

These figures add the file size, an fp16 KV cache at an 8K context window and about 0.5 GB for the runtime. They are estimates, so leave a little headroom, and expect to use less context or a smaller quant if you are right at the limit.

  • 4 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).
  • 48 GB VRAM: the best quant that fits is the main file (2.0 GB file, about 2.7 GB in use).

Which Step 3.5 Flash MTP quant should you choose?

IQ3_XXS (81.0 GB) is the middle option in this repository. A reasonable choice when a 4-bit file will not fit, mostly on larger models. Quality loss is noticeable but often acceptable.

If memory is tight, the smallest file is the first one in the table at 2.0 GB, with the quality trade-offs that come with fewer bits.

New to quant names? Read GGUF quantization explained.

Specifications

Architecture from the GGUF header

Architecturestep35
Parameters1.87B (approximate, mixture of experts)
Layers46
Trained context262,144 tokens

Source

Quantized bytnhnyzc
FamilyStep
Licenseapache-2.0
Downloads (HF)74
Last updated2026-06-03

Frequently asked questions about Step 3.5 Flash MTP

How much VRAM does Step 3.5 Flash MTP need?

About 91.2 GB at IQ3_XXS with an 8K context window. That is the 81.0 GB file, plus the KV cache and roughly 0.5 GB of runtime overhead. Smaller quants need less and longer context needs more.

Can I run Step 3.5 Flash MTP on an 8 GB GPU?

Yes, with the smaller files: the 2.0 GB file needs about 2.7 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Step 3.5 Flash MTP GGUF?

It depends on the quant: file is 2.0 GB, IQ3_XXS is 81.0 GB, IQ3_S is 89.9 GB, IQ4_XS is 96.3 GB.

What context length does Step 3.5 Flash MTP support?

The GGUF header lists a trained context length of 262,144 tokens (256K).

Where can I download Step 3.5 Flash MTP GGUF?

The files are on Hugging Face in the tnhnyzc/Step-3.5-Flash-MTP-GGUF repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.

What license does Step 3.5 Flash MTP use?

Hugging Face lists the license as apache-2.0. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.

Other GGUFs of Step-3.5-Flash

Similar models

Browse all →

More from tnhnyzc

All by tnhnyzc →

More Step models

All Step →

Other tiny models

All tiny models →