Model comparison
whisper-large-v3-turbo vs parakeet-unified-en-0.6b
At Q4_K_M, parakeet-unified-en-0.6b is the smaller download — 477,274,496 bytes against 536,069,728.
From the file· summed bytes, KV per layer
Side by side
| whisper-large-v3-turbo | parakeet-unified-en-0.6b | |
|---|---|---|
| Parameters | 809M | 618M |
| Architecture | whisper | parakeet |
| Layers | 32 | — |
| Native context | — | — |
| Mixture of experts | no | no |
| Quantizations published | 31 | 6 |
| Smallest quantization | 0.27 GiB | 0.44 GiB |
| Q4_K_M | 0.50 GiB | 0.44 GiB |
| Licence | apache-2.0 | cc-by-4.0 |
KV cache by context
the term that decides long-context viability
| Context | whisper-large-v3-turbo | parakeet-unified-en-0.6b | Ratio |
|---|---|---|---|
| 4,096 | — | — | — |
| 8,192 | — | — | — |
| 16,384 | — | — | — |
| 32,768 | — | — | — |
| 65,536 | — | — | — |
| 131,072 | — | — | — |