Model comparison
SciPhi-Self-RAG-Mistral-7B-32k vs Qwen3-4B
At Q4_K_M, Qwen3-4B is the smaller download — 2,497,280,256 bytes against 4,368,530,080.
From the file· summed bytes, KV per layer
Side by side
| SciPhi-Self-RAG-Mistral-7B-32k | Qwen3-4B | |
|---|---|---|
| Parameters | 7.2B | 4.0B |
| Architecture | llama | qwen3 |
| Layers | 32 | 36 |
| Native context | 32,768 | 40,960 |
| Mixture of experts | no | no |
| Quantizations published | 33 | 29 |
| Smallest quantization | 1.50 GiB | 1.01 GiB |
| Q4_K_M | 4.07 GiB | 2.33 GiB |
| Licence | mit | — |
KV cache by context
the term that decides long-context viability
| Context | SciPhi-Self-RAG-Mistral-7B-32k | Qwen3-4B | Ratio |
|---|---|---|---|
| 4,096 | — | 0.56 GiB | — |
| 8,192 | — | 1.13 GiB | — |
| 16,384 | — | 2.25 GiB | — |
| 32,768 | — | 4.50 GiB | — |
| 65,536 | — | 9.00 GiB | — |
| 131,072 | — | 18.00 GiB | — |