POCKET EN 160 needs about 6.5 GB of memory at its main file (a 5.3 GB file) with an 8K context window, so it fits on a 8 GB GPU. This repository has 2 quants, from 5.3 GB to 6.8 GB. Longer context needs more: at 32K tokens the KV cache alone adds about 2.7 GB.
POCKET EN 160 – 22.6B‑parameter Qwen3.6‑35B‑A3B MoE model for on‑device use
POCKET EN 160 is a 22.6‑billion‑parameter mixture‑of‑experts model based on Darwin‑36B‑Opus. It is quantized for GGUF and runs on iPhone and CPU‑only PCs. The model supports reasoning and tool usage.
Good at
- Reasoning tasks
- Tool integration
- On‑device inference
Limits
- No vision support
- Requires GGUF runtime
- Large memory footprint
Best for
- Mobile chat applications
- CPU‑only inference
- Low‑latency reasoning
Who it’s for. Developers building lightweight LLM apps on iOS or PCs without GPUs. The model is suitable for projects needing reasoning capabilities in a small footprint.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| — | 5.3 GB | 6.5 GB | Fits | Get file → | |
| — | 6.8 GB | 8.0 GB | Tight | Get file → |
The 5.33 GB file uses 6.5 GB of RAM at 8 k context. The 6.82 GB file uses 8 GB of RAM. For 8 GB VRAM systems, the 6.82 GB build is recommended; for 6 GB VRAM, the 5.33 GB build is safer. Estimates are approximate.
What GPU do you need for POCKET EN 160?
These figures add the file size, an fp16 KV cache at an 8K context window (worked out from this model’s own header) 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: the best quant that fits is the main file (6.8 GB file, about 8.0 GB in use).
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about POCKET EN 160
How much VRAM does POCKET EN 160 need?
About 6.5 GB at its main file with an 8K context window. That is the 5.3 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 POCKET EN 160 on an 8 GB GPU?
Yes, with the smaller files: the 6.8 GB file needs about 8.0 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of POCKET EN 160 GGUF?
It depends on the quant: file is 5.3 GB, file is 6.8 GB.
What context length does POCKET EN 160 support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.7 GB at 8K and 2.7 GB at 32K.
Where can I download POCKET EN 160 GGUF?
The files are on Hugging Face in the FINAL-Bench/POCKET-EN-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 POCKET EN 160 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.
Can I run POCKET EN 160 on a laptop with 8 GB VRAM?
Yes, the 6.82 GB GGUF file is designed for 8 GB VRAM and uses about 8 GB of RAM at 8 k context.
Does POCKET EN 160 support vision tasks?
No, the model has no vision capabilities.