Skip to content

feat(WO87): final MECE target taxonomy — 29 evidence-backed renames #362

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

Merged
ronaldtse merged 2 commits into from
Oct 24, 2026
Merged

feat(WO87): final MECE target taxonomy — 29 evidence-backed renames #362

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e7788bf
feat(WO87): final MECE target taxonomy — 29 renames applied zoo-wide
Oct 24, 2026
cfe27de
feat(WO87): sweep old ids across code, tests, README — filename field…
Oct 24, 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
38 changes: 19 additions & 19 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -87,20 +87,20 @@ CER unless stated).

| model | task | size | headline metric |
|---|---|---|---|
| `ara-diac-2.0` | Arabic diacritization | 2.78 GB fp32 | SadeedDiac-25 DER **2.29%** (beats GLM-5.2 2.69); WikiNews-2024 17.38 WER |
| ~~`ara-diac-news-1.1`~~ | Arabic news register | — | **WITHDRAWN** — artifact lost (never shipped; WO87). `ara-diac-2.0` covers the register |
| `ara-diac-small-2.1` | Arabic, fast tier | 1.42 GB fp32 | fullset DER 4.57 (teacher 2.29) |
| `ara-diac-plane-large-1.0` | Arabic, int8 fast tier | 0.87 GB int8 | DER 2.74 / WER 8.61 |
| `urd-diac-base-1.0` | Urdu diacritization | 1.35 GB fp16 | CER **3.57%** |
| `urd-diac-1.0` | Urdu, small tier | 1.42 GB fp32 | CER 3.74% |
| `urd-g2p-base-1.0` | Urdu G2P | 2.16 GB fp32 | CER 15.65% — **RESEARCH-ONLY** (non-commercial data) |
| `fas-diac-base-1.0` | Persian diacritization | 0.71 GB fp16 | CER **0.603%** row-mean / 0.777% corpus, 82.3% word-acc (CC0 data) |
| `fas-g2p-1.0` | Persian G2P | 2.78 GB fp32 | CER 1.6% |
| `tha-g2p-base-1.0` | Thai G2P | 2.78 GB fp32 | teacher PER 1.25% |
| `tha-g2p-small-1.0` | Thai G2P, int8 | 0.26 GB int8 | PER 2.85% (beam-4 12.06%) |
| `ara-vocalized-2.0` | Arabic diacritization | 2.78 GB fp32 | SadeedDiac-25 DER **2.29%** (beats GLM-5.2 2.69); WikiNews-2024 17.38 WER |
| ~~`ara-diac-news-1.1`~~ | Arabic news register | — | **WITHDRAWN** — artifact lost (never shipped; WO87). `ara-vocalized-2.0` covers the register |
| `ara-vocalized-small-2.1` | Arabic, fast tier | 1.42 GB fp32 | fullset DER 4.57 (teacher 2.29) |
| `ara-vocalized-plane-large-1.0` | Arabic, int8 fast tier | 0.87 GB int8 | DER 2.74 / WER 8.61 |
| `urd-vocalized-base-1.0` | Urdu diacritization | 1.35 GB fp16 | CER **3.57%** |
| `urd-vocalized-1.0` | Urdu, small tier | 1.42 GB fp32 | CER 3.74% |
| `urd-ipa-base-1.0` | Urdu G2P | 2.16 GB fp32 | CER 15.65% — **RESEARCH-ONLY** (non-commercial data) |
| `fas-vocalized-base-1.0` | Persian diacritization | 0.71 GB fp16 | CER **0.603%** row-mean / 0.777% corpus, 82.3% word-acc (CC0 data) |
| `fas-phonemic-1.0` | Persian G2P | 2.78 GB fp32 | CER 1.6% |
| `tha-phonemic-base-1.0` | Thai G2P | 2.78 GB fp32 | teacher PER 1.25% |
| `tha-phonemic-small-1.0` | Thai G2P, int8 | 0.26 GB int8 | PER 2.85% (beam-4 12.06%) |
| `khm-latn-1.0` | Khmer → Latin | 1.42 GB fp32 | CER 27.42% / EM 59.66% |
| `heb-diac-plane-2.0` | Hebrew nikud, int8 | 0.42 GB int8 | DER 8.18% |
| `heb-diac-1.1` | Hebrew nikud, fp32 | 2.78 GB fp32 | DER 16.44% |
| `heb-vocalized-plane-2.0` | Hebrew nikud, int8 | 0.42 GB int8 | DER 8.18% |
| `heb-vocalized-1.1` | Hebrew nikud, fp32 | 2.78 GB fp32 | DER 16.44% |

Quantized variants (int8/int4/fp16) ship as assets of their base
model's release. Hebrew remains data-supply-limited: newer lanes are
Expand Down Expand Up @@ -128,12 +128,12 @@ models above. Python (`SECRYST_INDEX`) and ruby

```python
from interscript_ml import Model # reference runtime (this repo)
m = Model.load("urd-diac-base-1.0") # resolves via the index, verifies sha256
m = Model.load("urd-vocalized-base-1.0") # resolves via the index, verifies sha256
print(m.translate("کلیدهای پرکینزمانند"))
```

Artifact notice (2026-10-21): `urd-diac-base-1.0` and
`fas-diac-base-1.0` were re-published with exporter-side fp16→fp32
Artifact notice (2026-10-21): `urd-vocalized-base-1.0` and
`fas-vocalized-base-1.0` were re-published with exporter-side fp16→fp32
graph-IO normalization — same weights, portable graph boundary (node
ONNX Runtime no longer corrupts fp16 bytes; the ruby loader, which
lacks Float16 materialization, decodes them). The re-published digests
Expand All @@ -145,10 +145,10 @@ sha256 verification and must be re-fetched.
In training, not shipped: the Arabic r8 noisy-student chain (stage 2),
the Khmer lexicon-conditioned arm, and the Thai lexicon-first arm —
each ships only after its stage-3 external-benchmark verdict clears
the gates. `urd-g2p-base-1.0` ships but is **research-only**
the gates. `urd-ipa-base-1.0` ships but is **research-only**
(non-commercial training data). `models.yaml` also carries
legacy/superseded generations (e.g. `ara-diac-1.0`, `layerdrop`,
`ara-diac-small-1.0/2.0`) that resolve via the index but are not
legacy/superseded generations (e.g. `ara-vocalized-1.0`, `layerdrop`,
`ara-vocalized-small-1.0/2.0`) that resolve via the index but are not
recommended for new use; the table above is the recommended set.

## Repositories
Expand Down
Loading
Loading