RLX compile bundle for
KittenML/kitten-tts-mini-0.8. Replaces ONNX Runtime at inference time:
rlx-kittentts loads
graph.json +
weights.safetensors, lowers to RLX HIR, and compiles with
rlx-runtime.
1# From rlx-models root
2just fetch-kittentts
3just fetch-kitten-rlx-bundle
4just kittentts -- --native --ipa "həˈloʊ" --voice Jasper --out-wav out.wav
1# 1. Upstream ONNX checkpoint + voices
2hf download KittenML/kitten-tts-mini-0.8 --local-dir .cache/kittentts-mini-0.8
3
4# 2. This RLX bundle (into rlx_bundle/ subdirectory)
5hf download eugenehp/kitten-tts-mini-0.8-rlx --local-dir .cache/kittentts-mini-0.8/rlx_bundle
6
7# 3. Sentinel file for native auto-detection
8ln -sf rlx_bundle/weights.safetensors .cache/kittentts-mini-0.8/model.safetensors
9
10# 4. Synthesize (from rlx-models checkout)
11cargo run -p rlx-kittentts --features native --release -- \
12 --model-dir .cache/kittentts-mini-0.8 \
13 --native --ipa "həˈloʊ" --voice Jasper --out-wav out.wav
1.cache/kittentts-mini-0.8/
2 config.json # from KittenML
3 voices.npz # from KittenML
4 kitten_tts_mini_v0_8.onnx
5 model.safetensors # symlink → rlx_bundle/weights.safetensors
6 rlx_bundle/ # this repo
7 graph.json
8 manifest.json
9 weights.safetensors
1just fetch-kittentts
2just export-kitten-rlx-bundle
Derived from
KittenML/kitten-tts-mini-0.8 — follow the upstream model license. RLX tooling (
rlx-models) is GPL-3.0.