Model comparison
granite-speech-4.1-2b vs parakeet-unified-en-0.6b
At Q4_K_M, parakeet-unified-en-0.6b is the smaller download — 477,274,496 bytes against 1,139,247,200.
From the file· summed bytes, KV per layer
Side by side
| granite-speech-4.1-2b | parakeet-unified-en-0.6b | |
|---|---|---|
| Parameters | 2.3B | 618M |
| Architecture | granite_speech | parakeet |
| Layers | 40 | — |
| Native context | 4,096 | — |
| Mixture of experts | no | no |
| Quantizations published | 13 | 6 |
| Smallest quantization | 1.06 GiB | 0.44 GiB |
| Q4_K_M | 1.06 GiB | 0.44 GiB |
| Licence | apache-2.0 | cc-by-4.0 |
KV cache by context
the term that decides long-context viability
| Context | granite-speech-4.1-2b | parakeet-unified-en-0.6b | Ratio |
|---|---|---|---|
| 4,096 | 0.31 GiB | — | — |
| 8,192 | 0.63 GiB | — | — |
| 16,384 | 1.25 GiB | — | — |
| 32,768 | 2.50 GiB | — | — |
| 65,536 | 5.00 GiB | — | — |
| 131,072 | 10.00 GiB | — | — |