Step 3.5 Flash REAP 149B A11B i1 needs about 101.2 GB of memory at Q4_K_M (a 89.9 GB file) with an 8K context window, which is more than a 48 GB GPU holds. This repository has 24 quants, from 0.4 GB to 122.7 GB (Q6_K).
Step‑3.5‑Flash‑REAP‑149B‑A11B – a 149B sparse MoE model with 11B active experts
Step‑3.5‑Flash‑REAP‑149B‑A11B is a 149B parameter sparse mixture‑of‑experts model. It uses REAP pruning to keep 11B active parameters per token and supports a 262,144 token context length. The model is compatible with vanilla vLLM and is distributed in GGUF format.
Good at
- Code generation
- Agentic coding
- Tool calling
Limits
- Large VRAM requirement for high‑quality quants
- No vision support
- License is other
Best for
- Local deployment on high‑memory GPUs
- Academic research on sparse MoE
- Resource‑constrained inference
Who it’s for. Developers and researchers needing a memory‑efficient version of Step‑3.5‑Flash with near‑lossless performance. The model is suitable for environments where 149B parameters are acceptable but only 11B are active per token.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 0.4 GB | 0.9 GB | Fits | Get file → | |
| IQ1_S | 30.0 GB | 34.1 GB | Too big | Get file → | |
| IQ1_M | 33.4 GB | 37.9 GB | Too big | Get file → | |
| IQ2_XXS | 39.1 GB | 44.3 GB | Too big | Get file → | |
| IQ2_XS | 43.7 GB | 49.4 GB | Too big | Get file → | |
| IQ2_S | 44.1 GB | 49.9 GB | Too big | Get file → | |
| IQ2_M | 48.7 GB | 55.0 GB | Too big | Get file → | |
| Q2_K_S | 50.0 GB | 56.5 GB | Too big | Get file → | |
| Q2_K | 54.5 GB | 61.6 GB | Too big | Get file → | |
| IQ3_XXS | 57.5 GB | 64.9 GB | Too big | Get file → | |
| IQ3_XS | 60.8 GB | 68.6 GB | Too big | Get file → | |
| Q3_K_S | 64.5 GB | 72.8 GB | Too big | Get file → | |
| IQ3_S | 64.6 GB | 72.8 GB | Too big | Get file → | |
| IQ3_M | 65.2 GB | 73.5 GB | Too big | Get file → | |
| Q3_K_M | 71.2 GB | 80.3 GB | Too big | Get file → | |
| Q3_K_L | 77.5 GB | 87.3 GB | Too big | Get file → | |
| IQ4_XS | 79.7 GB | 89.7 GB | Too big | Get file → | |
| Q4_0 | 84.4 GB | 95.1 GB | Too big | Get file → | |
| Q4_K_S | 84.6 GB | 95.3 GB | Too big | Get file → | |
| Q4_K_M | 89.9 GB | 101.2 GB | Too big | Get file → | |
| Q4_1 | 93.6 GB | 105.3 GB | Too big | Get file → | |
| Q5_K_S | 102.9 GB | 115.7 GB | Too big | Get file → | |
| Q5_K_M | 105.8 GB | 119.0 GB | Too big | Get file → | |
| Q6_K | 122.7 GB | 137.9 GB | Too big | Get file → |
For 8 GB VRAM, use the IQ1_S quant (30.02 GB file, 34.1 GB memory at 8k context). For 16 GB VRAM, IQ2_S (43.66 GB file, 49.4 GB memory) or Q2_K_S (50.03 GB file, 56.5 GB memory) are suitable. Higher‑quality quants such as Q4_K_M require 89.93 GB file and 101.2 GB memory.
What GPU do you need for Step 3.5 Flash REAP 149B A11B i1?
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 Step 3.5 Flash REAP 149B A11B i1 quant should you choose?
Q4_K_M is the usual starting point for Step 3.5 Flash REAP 149B A11B i1: it balances file size and quality well. Here it is a 89.9 GB file.
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
Source
Frequently asked questions about Step 3.5 Flash REAP 149B A11B i1
How much VRAM does Step 3.5 Flash REAP 149B A11B i1 need?
About 101.2 GB at Q4_K_M with an 8K context window. That is the 89.9 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 REAP 149B A11B i1 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 Step 3.5 Flash REAP 149B A11B i1 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 18 more.
What context length does Step 3.5 Flash REAP 149B A11B i1 support?
The GGUF header lists a trained context length of 262,144 tokens (256K).
Where can I download Step 3.5 Flash REAP 149B A11B i1 GGUF?
The files are on Hugging Face in the mradermacher/Step-3.5-Flash-REAP-149B-A11B-i1-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 REAP 149B A11B i1 use?
Hugging Face lists the license as other. Licenses can differ between the quantized files and the original model, so check the repository and the base model before commercial use.
Can I run this model on 8 GB VRAM?
Use the IQ1_S quant; it requires 34.1 GB memory at 8k context, which is above 8 GB but can be run with lower batch sizes.
Does the model support vision tasks?
No, the model has no vision capabilities.
What is the context length of this model?
The model supports a context length of 262,144 tokens.