Skip to content

parakeet-mlx: bf16 precision, native load, distributable bf16 model, download progress + dedup #8

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
fiorelorenzo wants to merge 15 commits into
base: main
Choose a base branch
from
Open

parakeet-mlx: bf16 precision, native load, distributable bf16 model, download progress + dedup #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
f6d7399
feat: streaming transcribe API for Parakeet ONNX engine
fiorelorenzo May 26, 2026
d63cf3a
refactor: re-decode full buffer per push so streaming matches one-shot
fiorelorenzo May 26, 2026
07388df
feat: streaming transcribe API for Parakeet MLX engine
fiorelorenzo May 26, 2026
d4ef5d7
bump ort to 2.0.0-rc.12
fiorelorenzo May 26, 2026
75ea39f
fix(parakeet-mlx): bound MLX Metal buffer cache during streaming
fiorelorenzo Jun 3, 2026
bb58823
feat(parakeet-mlx): load weights in bf16 to halve resident memory
fiorelorenzo Jun 3, 2026
549c2a2
fix(parakeet-mlx): down-cast attention weights + make forward reads b…
fiorelorenzo Jun 3, 2026
ee230f5
feat(parakeet-mlx): add ParakeetPrecision load param (int8 falls back…
fiorelorenzo Jun 3, 2026
ae35e13
perf(parakeet-mlx): eval bf16 weights at load so the fp32 copy frees …
fiorelorenzo Jun 3, 2026
d7004d6
feat(parakeet-mlx): load bf16/f16 weights natively without an f32 rou…
fiorelorenzo Jun 3, 2026
944408a
feat(examples): add convert_to_bf16 (fp32 safetensors -> bf16 for dis…
fiorelorenzo Jun 3, 2026
940d983
feat(parakeet-mlx): add parakeet-mlx arm to Model::from_dir + parity_…
fiorelorenzo Jun 3, 2026
3538b17
feat(parakeet-mlx): allow overriding the HF repo via AUDIOPIPE_PARAKE…
fiorelorenzo Jun 3, 2026
6869589
fix(model): dedup in-flight spawn_pretrained_download
fiorelorenzo Jun 3, 2026
5fb8a00
feat(parakeet-mlx): download_with_progress for real download progress…
fiorelorenzo Jun 3, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
174 changes: 29 additions & 145 deletions Cargo.lock
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

16 changes: 13 additions & 3 deletions Cargo.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -9,10 +9,10 @@ keywords = ["speech-to-text", "asr", "whisper", "parakeet", "onnx"]
categories = ["multimedia::audio", "science"]

[dependencies]
ort = { version = "=2.0.0-rc.10", default-features = false, features = ["ndarray", "half"], optional = true }
ort-sys = { version = "=2.0.0-rc.10", default-features = false, optional = true }
ort = { version = "=2.0.0-rc.12", default-features = false, features = ["ndarray", "half"], optional = true }
ort-sys = { version = "=2.0.0-rc.12", default-features = false, optional = true }
hf-hub = "0.4"
ndarray = { version = "0.16", optional = true }
ndarray = { version = "0.17", optional = true }
realfft = { version = "3", optional = true }
thiserror = "2"
tracing = "0.1"
Expand Down Expand Up @@ -47,6 +47,16 @@ path = "examples/bench_batch.rs"
name = "bench_asr"
path = "examples/bench_asr.rs"

[[example]]
name = "convert_to_bf16"
path = "examples/convert_to_bf16.rs"
required-features = ["parakeet-mlx"]

[[example]]
name = "parity_check"
path = "examples/parity_check.rs"
required-features = ["parakeet-mlx"]

[features]
default = ["ort-defaults"]

Expand Down
Loading