Step 3.5 Flash needs about 157.4 GB of memory at Q5_K (a 140.1 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 7 quants, from 72.4 GB (Q2_K) to 394.0 GB (BF16).
Step 3.5 Flash is a 197B-parameter Step model in GGUF format, quantized by milkowski. Its trained context length is 256K tokens. The repository has 7 quant files, from 72.4 GB (Q2_K) to 394.0 GB (BF16).
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q2_K | 72.4 GB | 81.6 GB | Too big | Get file → | |
| Q3_K | 94.4 GB | 106.3 GB | Too big | Get file → | |
| Q4_K | 119.2 GB | 134.0 GB | Too big | Get file → | |
| Q5_K | 140.1 GB | 157.4 GB | Too big | Get file → | |
| Q6_K | 162.4 GB | 182.3 GB | Too big | Get file → | |
| Q8_0 | 209.9 GB | 235.6 GB | Too big | Get file → | |
| BF16 | 394.0 GB | 441.8 GB | Too big | Get file → |
What GPU do you need for Step 3.5 Flash?
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: no quant fits at an 8K context.
- 6 GB VRAM: no quant fits at an 8K context.
- 8 GB VRAM: no quant fits at an 8K context.
- 12 GB VRAM: no quant fits at an 8K context.
- 16 GB VRAM: no quant fits at an 8K context.
- 24 GB VRAM: no quant fits at an 8K context.
- 32 GB VRAM: no quant fits at an 8K context.
- 48 GB VRAM: no quant fits at an 8K context.
Which Step 3.5 Flash quant should you choose?
Q5_K (140.1 GB) is the middle option in this repository. Closer to the original model than 4-bit, for a moderate increase in size. A good pick when you have memory to spare.
If memory is tight, the smallest file is Q2_K at 72.4 GB, with the quality trade-offs that come with fewer bits. Use it only when memory is very tight. Expect a clear drop in quality, which is usually worse on small models and easier to live with on large ones.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about Step 3.5 Flash
How much VRAM does Step 3.5 Flash need?
About 157.4 GB at Q5_K with an 8K context window. That is the 140.1 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 on an 8 GB GPU?
Not at an 8K context. The smallest file needs about 81.6 GB, so you would need a larger GPU or CPU offloading.
What is the file size of Step 3.5 Flash GGUF?
It depends on the quant: Q2_K is 72.4 GB, Q3_K is 94.4 GB, Q4_K is 119.2 GB, Q5_K is 140.1 GB, Q6_K is 162.4 GB, Q8_0 is 209.9 GB and 1 more.
What context length does Step 3.5 Flash support?
The GGUF header lists a trained context length of 262,144 tokens (256K).
Where can I download Step 3.5 Flash GGUF?
The files are on Hugging Face in the milkowski/Step-3.5-Flash-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 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.