SatellaRAG 0.8B needs about 1.5 GB of memory at Q4_K_M (a 0.5 GB file) with an 8K context window, so it fits on a 4 GB GPU. This repository has 5 quants, from 0.5 GB (Q4_K_M) to 1.6 GB (F16). Longer context needs more: at 32K tokens the KV cache alone adds about 1.7 GB.
SatellaRAG 0.8B – a retrieval‑grounded reasoning model
SatellaRAG 0.8B is a fine‑tuned version of Qwen3.5‑0.8B, optimized for retrieval‑augmented generation. It prioritizes supplied evidence, preserves citations, and signals when evidence is insufficient. The vision parameters are frozen and the model is text‑focused.
Good at
- Grounded question answering
- Evidence prioritization
- Citation preservation
Limits
- No vision inference
- Requires external retrieval pipeline
- Limited to 262144 context tokens
Best for
- RAG pipelines
- Document‑centric QA
- Citation‑aware generation
Who it’s for. Developers building retrieval‑augmented applications who need a lightweight, text‑only generation component. Researchers prototyping grounded reasoning models may also benefit.
Quants & memory
| Quant | File | Est. memory | vs VRAM | Result | |
|---|---|---|---|---|---|
| Q4_K_M | 0.5 GB | 1.5 GB | Fits | Get file → | |
| Q5_K_M | 0.6 GB | 1.5 GB | Fits | Get file → | |
| Q6_K | 0.6 GB | 1.6 GB | Fits | Get file → | |
| Q8_0 | 0.8 GB | 1.8 GB | Fits | Get file → | |
| F16 | 1.6 GB | 2.5 GB | Fits | Get file → |
For 8 GB VRAM use Q4_K_M (0.54 GB, ~1.5 GB memory). Q5_K_M (0.59 GB) offers similar size with slightly higher quality. For higher fidelity use Q8_0 (0.83 GB, ~1.8 GB memory).
What GPU do you need for SatellaRAG 0.8B?
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 F16 (1.6 GB file, about 2.5 GB in use).
Which SatellaRAG 0.8B quant should you choose?
Q4_K_M is the usual starting point for SatellaRAG 0.8B: it balances file size and quality well. Here it is a 0.5 GB file.
The Q8_0 file is 0.8 GB, about 1.5 times the size of Q4_K_M. It keeps the weights closer to the original, which only matters if you have the memory to spare.
New to quant names? Read GGUF quantization explained.
Specifications
Architecture from the GGUF header
Source
Frequently asked questions about SatellaRAG 0.8B
How much VRAM does SatellaRAG 0.8B need?
About 1.5 GB at Q4_K_M with an 8K context window. That is the 0.5 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 SatellaRAG 0.8B on an 8 GB GPU?
Yes, with F16: the 1.6 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 SatellaRAG 0.8B GGUF?
It depends on the quant: Q4_K_M is 0.5 GB, Q5_K_M is 0.6 GB, Q6_K is 0.6 GB, Q8_0 is 0.8 GB, F16 is 1.6 GB.
What context length does SatellaRAG 0.8B support?
The GGUF header lists a trained context length of 262,144 tokens (256K). The KV cache grows with the context you use: about 0.4 GB at 8K and 1.7 GB at 32K.
Where can I download SatellaRAG 0.8B GGUF?
The files are on Hugging Face in the EphAsad/SatellaRAG-0.8B repository. The download buttons on this page open the file on Hugging Face. ggufmodels.com does not host any model files.
What license does SatellaRAG 0.8B 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 SatellaRAG on 8 GB VRAM?
Yes, Q4_K_M or Q5_K_M quantizations fit within 8 GB VRAM with ~1.5 GB memory usage at 8 k context.
Does SatellaRAG support vision tasks?
No, the vision parameters were frozen during training and the model is text‑focused.
What license applies to SatellaRAG?
The model is released under the apache-2.0 license.