Downloads · 30 days
14
9% of all-time downloads
nllb200-coreml-256-ane
aoiandroid/nllb200-coreml-256-ane
nllb200-coreml-256-ane is a translation model from aoiandroid. Use it when you need text moved from one language to another. It is set up for coreml. The card lists the license as cc-by-nc-4.0.
Core ML .mlpackage bundles for encoder and KV-cache decoder (init + step), exported from facebook/nllb-200-distilled-600M. Fixed sequence length 256, ANE-oriented graph, float32 weights (no palettization in this revis…
All-time downloads
150
Public
Repo size
4.5 GB
Likes
0
Public
Repo makeup
Click a slice to open those files.
.bin4.5 GB · 99%
At a glance
- Task
- Translation
- Library
- coreml
- License
- cc-by-nc-4.0
- Model type
- m2m_100
- Access
- Public
- Created
- Mar 27, 2026
- Updated
- Mar 27, 2026
- SHA
- d8855d9d
Base models
Model card
From the Hugging Face model README
NLLB-200 distilled 600M — Core ML (seq 256, ANE, float32)
Core ML .mlpackage bundles for encoder and KV-cache decoder (init + step), exported from facebook/nllb-200-distilled-600M. Fixed sequence length 256, ANE-oriented graph, float32 weights (no palettization in this revision).
Contents
| Artifact | Role |
|---|---|
NLLB_Encoder_256.mlpackage | Encode source tokens |
NLLB_Decoder_256_init.mlpackage | Decoder first step (no past KV) |
NLLB_Decoder_256_step.mlpackage | Autoregressive step with past KV |
config.json | Architecture metadata (seq_len, compute_units, etc.) |
tokenizer/ | Reference tokenizer files for the app pipeline |
Companion variant
- Palettized (pal8) weights (smaller on disk, different quality/latency tradeoffs):
aoiandroid/nllb200-coreml-256-ane-pal8
Config highlights (config.json)
model_type:m2m_100/M2M100ForConditionalGenerationseq_len: 256ane_optimized: truecompute_units: CPU_AND_NEdtype: float32
Base model and license
- Source weights: NLLB-200 distilled 600M (Meta / FAIR).
- Original terms are CC-BY-NC 4.0. This repository only ships a derived Core ML export for on-device use; the same license restrictions apply.
Intended use
- macOS / iOS on-device translation via
MLModel(orcoremltoolsin Python for validation). - Language direction is controlled in your app: tokenizer
src_lang, forced target BOS (e.g.eng_Latn), and your beam/greedy loop aroundDecoder_256_step. - Validate quality and performance on your target hardware (ANE vs CPU/GPU behavior can differ).
Limitations
- Fixed max sequence length 256 at export time; longer inputs require truncation or a different export.
- Not a drop-in Hugging Face
transformerscheckpoint; this is a split Core ML deployment (encoder + decoder packages).
Maintainer
- Task
- Translation
- Library
- coreml
- Type
- m2m_100
- License
- cc-by-nc-4.0
- Languages
- multilingual
- Created
- Mar 27, 2026
- Updated
- Mar 27, 2026