ST

Cerebras Step 3.5 Flash REAP 149B A11B

StepGGUF by wimmmmUpdated 2026-03-22
Download on Hugging Face
Parameters149B
Context256K
Layers—
Architecturestep35
Smallest quant0.4 GB
Quick answer

Cerebras Step 3.5 Flash REAP 149B A11B needs about 68.6 GB of memory at IQ3_XS (a 60.8 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 17 quants, from 0.4 GB to 298.8 GB (BF16).

Cerebras Step 3.5 Flash REAP 149B A11B is a 149B-parameter Step model in GGUF format, quantized by wimmmm. Its trained context length is 256K tokens. The repository has 17 quant files, from 0.4 GB to 298.8 GB (BF16).

Quants & memory

QuantFileEst. memoryvs VRAMResultDownload
—0.4 GB 0.9 GB
Fits Get file →
IQ1_S30.0 GB 34.1 GB
Too big Get file →
IQ1_M33.4 GB 37.9 GB
Too big Get file →
IQ2_XXS39.1 GB 44.3 GB
Too big Get file →
IQ2_XS43.7 GB 49.4 GB
Too big Get file →
IQ2_S44.1 GB 49.9 GB
Too big Get file →
IQ2_M48.7 GB 55.0 GB
Too big Get file →
IQ3_XXS57.5 GB 64.9 GB
Too big Get file →
IQ3_XS60.8 GB 68.6 GB
Too big Get file →
IQ3_S64.6 GB 72.8 GB
Too big Get file →
IQ3_M65.2 GB 73.5 GB
Too big Get file →
IQ4_XS79.7 GB 89.7 GB
Too big Get file →
IQ4_NL84.3 GB 94.9 GB
Too big Get file →
Q4_K89.9 GB 101.2 GB
Too big Get file →
Q5_K105.8 GB 119.0 GB
Too big Get file →
Q6_K122.7 GB 137.9 GB
Too big Get file →
BF16298.8 GB 335.2 GB
Too big Get file →

What GPU do you need for Cerebras Step 3.5 Flash REAP 149B A11B?

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 (0.4 GB file, about 0.9 GB in use).
  • 6 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 8 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 12 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 16 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 24 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 32 GB VRAM: the best quant that fits is the main file (0.4 GB file, about 0.9 GB in use).
  • 48 GB VRAM: the best quant that fits is IQ2_XXS (39.1 GB file, about 44.3 GB in use).

Which Cerebras Step 3.5 Flash REAP 149B A11B quant should you choose?

IQ3_XS (60.8 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 0.4 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
Parameters149B
Trained context262,144 tokens

Source

Quantized bywimmmm
FamilyStep
Licenseapache-2.0
Downloads (HF)824
Last updated2026-03-22

Frequently asked questions about Cerebras Step 3.5 Flash REAP 149B A11B

How much VRAM does Cerebras Step 3.5 Flash REAP 149B A11B need?

About 68.6 GB at IQ3_XS with an 8K context window. That is the 60.8 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 Cerebras Step 3.5 Flash REAP 149B A11B on an 8 GB GPU?

Yes, with the smaller files: the 0.4 GB file needs about 0.9 GB at 8K context. Larger quants in this repository do not fit in 8 GB.

What is the file size of Cerebras Step 3.5 Flash REAP 149B A11B GGUF?

It depends on the quant: file is 0.4 GB, IQ1_S is 30.0 GB, IQ1_M is 33.4 GB, IQ2_XXS is 39.1 GB, IQ2_XS is 43.7 GB, IQ2_S is 44.1 GB and 11 more.

What context length does Cerebras Step 3.5 Flash REAP 149B A11B support?

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

Where can I download Cerebras Step 3.5 Flash REAP 149B A11B GGUF?

The files are on Hugging Face in the wimmmm/Cerebras-Step-3.5-Flash-REAP-149B-A11B-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 Cerebras Step 3.5 Flash REAP 149B A11B 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-REAP-149B-A11B

Similar models

Browse all →

More from wimmmm

All by wimmmm →

More Step models

All Step →

Other large models

All large models →