Transformers
Safetensors
t5
text2text-generation
protein-language-model
fastplms
custom_code
text-generation-inference
Instructions to use Synthyra/ANKH_base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Synthyra/ANKH_base with Transformers:
# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForSeq2SeqLM tokenizer = AutoTokenizer.from_pretrained("Synthyra/ANKH_base", trust_remote_code=True) model = AutoModelForSeq2SeqLM.from_pretrained("Synthyra/ANKH_base", trust_remote_code=True, device_map="auto") - Notebooks
- Google Colab
- Kaggle
Update FastPLMs runtime files
Browse filesAdd-only FastPLMs files-only publication. Checkpoint weights and complete-artifact attestations are unchanged.
- README.md +4 -5
- config.json +3 -3
- runtime-attestation.json +5 -5
README.md
CHANGED
|
@@ -20,11 +20,11 @@ Supported Transformers entry points are `AutoConfig`, `AutoModel`,
|
|
| 20 |
|
| 21 |
## Install and platform requirements
|
| 22 |
|
| 23 |
-
Install
|
| 24 |
|
| 25 |
```bash
|
| 26 |
python -m pip install \
|
| 27 |
-
"fastplms @ git+https://github.com/Synthyra/FastPLMs.git
|
| 28 |
```
|
| 29 |
|
| 30 |
Python 3.11-3.14, PyTorch 2.13, and Transformers 5.13 are required. The declared CPU gate covers tiny offline contracts; published checkpoint throughput and parity require the documented device tier. The Hub quick start below requires network
|
|
@@ -186,9 +186,8 @@ masked-LM extension and is not an official ANKH head.
|
|
| 186 |
## Release record
|
| 187 |
|
| 188 |
- FastPLMs weights: `Synthyra/ANKH_base`
|
| 189 |
-
- Runtime revision:
|
| 190 |
-
-
|
| 191 |
-
- Runtime bundle SHA-256: `405e7dddef77117e55232e1112faf088db88bb9b13e212a63b3318c4bc0d5a03`
|
| 192 |
- Generator/schema version and complete/runtime-only attestations: recorded in `provenance.json`
|
| 193 |
- Canonical transformed state SHA-256: `cdd8d30d88e5bf41f44e1eef4470d8e46607aba5f7c7c805b06c035b89c8c16f`
|
| 194 |
- Conversion equality attestation: recorded in `provenance.json`
|
|
|
|
| 20 |
|
| 21 |
## Install and platform requirements
|
| 22 |
|
| 23 |
+
Install the current FastPLMs package:
|
| 24 |
|
| 25 |
```bash
|
| 26 |
python -m pip install \
|
| 27 |
+
"fastplms @ git+https://github.com/Synthyra/FastPLMs.git"
|
| 28 |
```
|
| 29 |
|
| 30 |
Python 3.11-3.14, PyTorch 2.13, and Transformers 5.13 are required. The declared CPU gate covers tiny offline contracts; published checkpoint throughput and parity require the documented device tier. The Hub quick start below requires network
|
|
|
|
| 186 |
## Release record
|
| 187 |
|
| 188 |
- FastPLMs weights: `Synthyra/ANKH_base`
|
| 189 |
+
- Runtime revision: recorded separately in the built artifact and published commit
|
| 190 |
+
- Source-tree and runtime-bundle SHA-256: recorded in `provenance.json`
|
|
|
|
| 191 |
- Generator/schema version and complete/runtime-only attestations: recorded in `provenance.json`
|
| 192 |
- Canonical transformed state SHA-256: `cdd8d30d88e5bf41f44e1eef4470d8e46607aba5f7c7c805b06c035b89c8c16f`
|
| 193 |
- Conversion equality attestation: recorded in `provenance.json`
|
config.json
CHANGED
|
@@ -22,10 +22,10 @@
|
|
| 22 |
"fastplms_checkpoint_repo_id": "ElnaggarLab/ankh-base",
|
| 23 |
"fastplms_checkpoint_revision": "d99cb6b966530dfc2ae96bc69d9255c2a07308b0",
|
| 24 |
"fastplms_model_id": "ankh_base",
|
| 25 |
-
"fastplms_release_tool_revision": "
|
| 26 |
-
"fastplms_release_tool_sha256": "
|
| 27 |
"fastplms_runtime_bundle_sha256": "405e7dddef77117e55232e1112faf088db88bb9b13e212a63b3318c4bc0d5a03",
|
| 28 |
-
"fastplms_runtime_revision": "
|
| 29 |
"fastplms_source_tree_sha256": "e377aef8902771bb8bdabcae9c8853ab50de04d85cacd3d75c6e83508f6d52ed",
|
| 30 |
"fastplms_weights_revision": "d99cb6b966530dfc2ae96bc69d9255c2a07308b0",
|
| 31 |
"feed_forward_proj": "gated-gelu",
|
|
|
|
| 22 |
"fastplms_checkpoint_repo_id": "ElnaggarLab/ankh-base",
|
| 23 |
"fastplms_checkpoint_revision": "d99cb6b966530dfc2ae96bc69d9255c2a07308b0",
|
| 24 |
"fastplms_model_id": "ankh_base",
|
| 25 |
+
"fastplms_release_tool_revision": "95691a87c781f052f65b6ab3ca99dfce4ebb59c7",
|
| 26 |
+
"fastplms_release_tool_sha256": "19b4247aba04a74de5069397518374ab2c3b9e0ce18ba9f36de34f0f4c5d1ceb",
|
| 27 |
"fastplms_runtime_bundle_sha256": "405e7dddef77117e55232e1112faf088db88bb9b13e212a63b3318c4bc0d5a03",
|
| 28 |
+
"fastplms_runtime_revision": "95691a87c781f052f65b6ab3ca99dfce4ebb59c7",
|
| 29 |
"fastplms_source_tree_sha256": "e377aef8902771bb8bdabcae9c8853ab50de04d85cacd3d75c6e83508f6d52ed",
|
| 30 |
"fastplms_weights_revision": "d99cb6b966530dfc2ae96bc69d9255c2a07308b0",
|
| 31 |
"feed_forward_proj": "gated-gelu",
|
runtime-attestation.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"files": {
|
| 3 |
"LICENSES/FastPLMs-Apache-2.0.txt": "sha256:2d2b50c7b1414bff1189a1db1f0cfb92e3e064b50f4c2b1019827b683e1b629a",
|
| 4 |
"LICENSES/ankh/LICENSE.md": "sha256:cd041d7f9f52936e8824ac3f754e9c67410763205fc8a7020ba74fc8b6edc088",
|
| 5 |
-
"README.md": "sha256:
|
| 6 |
"THIRD_PARTY_NOTICES.md": "sha256:25704b3c76404696cae52e7fca13088d329f70f412687340351259e86cd62baa",
|
| 7 |
-
"config.json": "sha256:
|
| 8 |
"fastplms/__init__.py": "sha256:4fb3196022ca8ec699d59d09bdbc5f0184195552b773698ab9b061fe3cd7df12",
|
| 9 |
"fastplms/attention/__init__.py": "sha256:f60b9fecfb4bcb37a4e7c26dc2f752b9035f9cbad627b4a84213f3a92ec88f7d",
|
| 10 |
"fastplms/attention/_core.py": "sha256:8f7ec5b65bd8b6c6fa4951d50d1c0e499abf03ae00914794b51fc410201e3e33",
|
|
@@ -30,10 +30,10 @@
|
|
| 30 |
},
|
| 31 |
"model_id": "ankh_base",
|
| 32 |
"redistributable": true,
|
| 33 |
-
"release_tool_revision": "
|
| 34 |
-
"release_tool_sha256": "
|
| 35 |
"runtime_bundle_sha256": "405e7dddef77117e55232e1112faf088db88bb9b13e212a63b3318c4bc0d5a03",
|
| 36 |
-
"runtime_revision": "
|
| 37 |
"schema_version": 2,
|
| 38 |
"scope": "runtime-only",
|
| 39 |
"source_tree_sha256": "e377aef8902771bb8bdabcae9c8853ab50de04d85cacd3d75c6e83508f6d52ed",
|
|
|
|
| 2 |
"files": {
|
| 3 |
"LICENSES/FastPLMs-Apache-2.0.txt": "sha256:2d2b50c7b1414bff1189a1db1f0cfb92e3e064b50f4c2b1019827b683e1b629a",
|
| 4 |
"LICENSES/ankh/LICENSE.md": "sha256:cd041d7f9f52936e8824ac3f754e9c67410763205fc8a7020ba74fc8b6edc088",
|
| 5 |
+
"README.md": "sha256:2fc528d8aac20cb5a161e77f4b9cfa2d789151967c083410088d3a865bd7fc68",
|
| 6 |
"THIRD_PARTY_NOTICES.md": "sha256:25704b3c76404696cae52e7fca13088d329f70f412687340351259e86cd62baa",
|
| 7 |
+
"config.json": "sha256:d867b0de3917e8b85a1b0069d8cae0519e4508bd85f79f38fbff9e15e11bda41",
|
| 8 |
"fastplms/__init__.py": "sha256:4fb3196022ca8ec699d59d09bdbc5f0184195552b773698ab9b061fe3cd7df12",
|
| 9 |
"fastplms/attention/__init__.py": "sha256:f60b9fecfb4bcb37a4e7c26dc2f752b9035f9cbad627b4a84213f3a92ec88f7d",
|
| 10 |
"fastplms/attention/_core.py": "sha256:8f7ec5b65bd8b6c6fa4951d50d1c0e499abf03ae00914794b51fc410201e3e33",
|
|
|
|
| 30 |
},
|
| 31 |
"model_id": "ankh_base",
|
| 32 |
"redistributable": true,
|
| 33 |
+
"release_tool_revision": "95691a87c781f052f65b6ab3ca99dfce4ebb59c7",
|
| 34 |
+
"release_tool_sha256": "19b4247aba04a74de5069397518374ab2c3b9e0ce18ba9f36de34f0f4c5d1ceb",
|
| 35 |
"runtime_bundle_sha256": "405e7dddef77117e55232e1112faf088db88bb9b13e212a63b3318c4bc0d5a03",
|
| 36 |
+
"runtime_revision": "95691a87c781f052f65b6ab3ca99dfce4ebb59c7",
|
| 37 |
"schema_version": 2,
|
| 38 |
"scope": "runtime-only",
|
| 39 |
"source_tree_sha256": "e377aef8902771bb8bdabcae9c8853ab50de04d85cacd3d75c6e83508f6d52ed",
|