Upload flan-t5-small ONNX model

README.md

@@ -0,0 +1,51 @@
+---
+library_name: onnx
+tags:
+  - text2text-generation
+  - t5
+  - flan-t5
+  - encoder-decoder
+  - onnx
+  - inference4j
+license: apache-2.0
+pipeline_tag: text2text-generation
+---
+
+# Flan-T5 Small — ONNX
+
+ONNX export of [Flan-T5 Small](https://huggingface.co/google/flan-t5-small) (77M parameters) with encoder-decoder architecture and KV cache support.
+
+Converted for use with [inference4j](https://github.com/inference4j/inference4j), an inference-only AI library for Java.
+
+## Original Source
+
+- **Repository:** [google/flan-t5-small](https://huggingface.co/google/flan-t5-small)
+- **License:** Apache 2.0
+
+## Usage with inference4j
+
+```java
+// Summarization
+try (var gen = FlanT5TextGenerator.flanT5Small().build()) {
+    System.out.println(gen.summarize("Long article text..."));
+}
+
+// Translation
+try (var gen = FlanT5TextGenerator.flanT5Small().build()) {
+    System.out.println(gen.translate("Hello!", Language.EN, Language.FR));
+}
+```
+
+## Model Details
+
+| Property | Value |
+|----------|-------|
+| Architecture | T5 encoder-decoder (77M parameters) |
+| Tasks | Summarization, translation, grammar correction, text-to-SQL |
+| Tokenizer | SentencePiece (32,128 tokens) |
+| Original framework | PyTorch (transformers) |
+| Export method | Hugging Face Optimum (encoder-decoder with KV cache) |
+
+## License
+
+This model is licensed under the [Apache License 2.0](https://www.apache.org/licenses/LICENSE-2.0). Original model by [Google](https://huggingface.co/google).

config.json

@@ -0,0 +1,61 @@
+{
+  "architectures": [
+    "T5ForConditionalGeneration"
+  ],
+  "classifier_dropout": 0.0,
+  "d_ff": 1024,
+  "d_kv": 64,
+  "d_model": 512,
+  "decoder_start_token_id": 0,
+  "dense_act_fn": "gelu_new",
+  "dropout_rate": 0.1,
+  "dtype": "float32",
+  "eos_token_id": 1,
+  "feed_forward_proj": "gated-gelu",
+  "initializer_factor": 1.0,
+  "is_encoder_decoder": true,
+  "is_gated_act": true,
+  "layer_norm_epsilon": 1e-06,
+  "model_type": "t5",
+  "n_positions": 512,
+  "num_decoder_layers": 8,
+  "num_heads": 6,
+  "num_layers": 8,
+  "output_past": true,
+  "pad_token_id": 0,
+  "relative_attention_max_distance": 128,
+  "relative_attention_num_buckets": 32,
+  "task_specific_params": {
+    "summarization": {
+      "early_stopping": true,
+      "length_penalty": 2.0,
+      "max_length": 200,
+      "min_length": 30,
+      "no_repeat_ngram_size": 3,
+      "num_beams": 4,
+      "prefix": "summarize: "
+    },
+    "translation_en_to_de": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to German: "
+    },
+    "translation_en_to_fr": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to French: "
+    },
+    "translation_en_to_ro": {
+      "early_stopping": true,
+      "max_length": 300,
+      "num_beams": 4,
+      "prefix": "translate English to Romanian: "
+    }
+  },
+  "tie_word_embeddings": false,
+  "transformers_version": "4.57.6",
+  "use_cache": true,
+  "vocab_size": 32128
+}

decoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f65e029938345ee2cfa2c059a3532c9b2f165cbd22d256b66206feb132d92f2
+size 232577642

decoder_with_past_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a2696cf3978727ccf37aaca27cc88c0f2ee017b4c79072ba9b51eb1c71c7e2bb
+size 219971365

encoder_model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8e01244ad371cb9405baa37f4c126e40c6b601637dcdddb8058d639ea638d21
+size 141463091