DeepNarra-Paham-Experimental needs about 0.9 GB of memory at TQ1_0 (a 0.3 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 2 quants, from 0.3 GB (TQ1_0) to 1.9 GB.
DeepNarra Paham-Experimental: 3‑Tier Awake‑Sleep continual learning engine
DeepNarra Paham‑Experimental is a research variant of DeepNarra/Paham designed for decoupled 3‑Tier Awake/Sleep continual learning. It uses recurrent fast‑weights, sidecar‑QLoRA, MoE router gating, and 75/25 replay consolidation. The model is licensed under Apache‑2.0.
Good at
- Continual learning with decoupled training phases
- MoE routing for dynamic inference
- Fast‑weight updates for rapid adaptation
Limits
- Context length limited to 4096 tokens
- No vision support
- Requires active user interaction for consolidation
Best for
- Research on continual learning architectures
- Edge models needing lifelong learning
- Prototyping MoE‑based systems
Who it’s for. Researchers and developers working on continual learning and edge AI. The model is suitable for experimental setups where decoupled training and inference are required.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| TQ1_0 | 0.3 GB | 0.8 GB | Fits | Get file → | |
| — | 1.9 GB | 2.4 GB | Fits | Get file → |
The TQ1_0 quantized file is 0.19 GB and requires approximately 0.8 GB of VRAM at 8K context. This is suitable for systems with at least 1 GB of VRAM. For larger context lengths, memory usage will increase proportionally.
What GPU do you need for DeepNarra-Paham-Experimental?
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: the best quant that fits is the main file (1.9 GB file, about 2.5 GB in use).
Which DeepNarra-Paham-Experimental quant should you choose?
TQ1_0 (0.3 GB) is the middle option in this repository. Lower bit counts mean smaller files and more quality loss.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about DeepNarra-Paham-Experimental
How much VRAM does DeepNarra-Paham-Experimental need?
About 0.9 GB at TQ1_0 with an 8K context window. That is the 0.3 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 DeepNarra-Paham-Experimental on an 8 GB GPU?
Yes, with the smaller files: the 1.9 GB file needs about 2.5 GB at 8K context. Larger quants in this repository do not fit in 8 GB.
What is the file size of DeepNarra-Paham-Experimental GGUF?
It depends on the quant: TQ1_0 is 0.3 GB, file is 1.9 GB.
What context length does DeepNarra-Paham-Experimental support?
The GGUF header lists a trained context length of 2,048 tokens (2K). The KV cache grows with the context you use: about 0.1 GB at 8K and 0.0 GB at 2K.
Where can I download DeepNarra-Paham-Experimental GGUF?
The files are on Hugging Face in the DeepNarra/Paham-Experimental repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does DeepNarra-Paham-Experimental 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 DeepNarra Paham-Experimental on 8 GB VRAM?
Yes. The TQ1_0 quant requires about 0.8 GB of VRAM at 8K context, well below 8 GB.
What is the maximum context length supported?
The model supports up to 4096 tokens of context.
Does the model support vision tasks?
No, the model does not include vision capabilities.