Run handy-computer/parakeet-unified-en-0.6b-gguf locally

License: cc-by-4.0 ⬇ 475,295 ❤ 1
Parameters0.62B

handy-computer/parakeet-unified-en-0.6b-gguf is a compact language model with 0.62 billion parameters, built on the parakeet architecture. It is released under the cc-by-4.0 license and has been downloaded 475,295 times.

To run handy-computer/parakeet-unified-en-0.6b-gguf locally at a 4,096-token context, its quantized versions need between 1.62 GB (Q4_K_M, lowest quality) and 3.48 GB (F32, highest quality) of memory, weights plus KV cache and a system margin included.

For most users the best balance is F32, needing about 3.48 GB. That means handy-computer/parakeet-unified-en-0.6b-gguf fits entirely in the VRAM of a 6 GB GPU or larger, running fully on the GPU.

→ Guide: How much VRAM do you need?

All quantizations

Quant.Bits QualityWeights KVTotal Speed~Verdict
Q4_K_M 6.18 Very good 0.44 GB 0.38 GB 1.62 GB 899.9 t/s Fits in VRAM
Q5_K_M 7.0 Excellent 0.5 GB 0.38 GB 1.68 GB 794.2 t/s Fits in VRAM
Q6_K 7.79 Excellent 0.56 GB 0.38 GB 1.74 GB 713.2 t/s Fits in VRAM
Q8_0 9.46 Excellent 0.68 GB 0.38 GB 1.86 GB 587.3 t/s Fits in VRAM
F16 16.03 Excellent 1.15 GB 0.38 GB 2.33 GB 346.6 t/s Fits in VRAM
F32 32.0 Excellent 2.3 GB 0.38 GB 3.48 GB 173.7 t/s Fits in VRAM

KV cache estimated (architecture unavailable). Speed is a rough estimate bounded by memory bandwidth.

Frequently asked questions

How much VRAM do you need to run handy-computer/parakeet-unified-en-0.6b-gguf?

You need about 3.48 GB of VRAM to run handy-computer/parakeet-unified-en-0.6b-gguf entirely on the GPU using the F32 quantization (at a 4,096-token context). Smaller quantizations lower the requirement at the cost of quality.

Can I run handy-computer/parakeet-unified-en-0.6b-gguf on an 8 GB GPU?

Yes. With 8 GB of VRAM you can run handy-computer/parakeet-unified-en-0.6b-gguf fully on the GPU using F32 (about 3.48 GB).

Can I run handy-computer/parakeet-unified-en-0.6b-gguf on a 16 GB GPU?

Yes. With 16 GB of VRAM you can run handy-computer/parakeet-unified-en-0.6b-gguf fully on the GPU using F32 (about 3.48 GB).

Can I run handy-computer/parakeet-unified-en-0.6b-gguf on a 24 GB GPU?

Yes. With 24 GB of VRAM you can run handy-computer/parakeet-unified-en-0.6b-gguf fully on the GPU using F32 (about 3.48 GB).

What is the best quantization for handy-computer/parakeet-unified-en-0.6b-gguf?

If memory allows, higher bits-per-weight means better quality. A common sweet spot is a Q4_K_M or Q5_K_M quantization, which keeps most of the quality while roughly halving the memory versus 8-bit. Pick the highest quantization that still fits in your VRAM.