Upload folder using huggingface_hub

README.md

@@ -0,0 +1,86 @@
+---
+language:
+- en
+license: apache-2.0
+pipeline_tag: text-generation
+tags:
+- transformers
+library_name: transformers.js
+base_model:
+- PleIAs/Monad
+---
+
+
+
+# Monad (ONNX)
+
+
+This is an ONNX version of [PleIAs/Monad](https://huggingface.co/PleIAs/Monad). It was automatically converted and uploaded using [this Hugging Face Space](https://huggingface.co/spaces/onnx-community/convert-to-onnx).
+
+
+## Usage with Transformers.js
+
+
+See the pipeline documentation for `text-generation`: https://huggingface.co/docs/transformers.js/api/pipelines#module_pipelines.TextGenerationPipeline
+
+
+---
+
+
+# ⚛️ Monad
+
+<div align="center">
+  <img src="figures/pleias.jpg" width="60%" alt="Pleias" />
+</div>
+
+<p align="center">
+  <a href="https://pleias.fr/blog/blogsynth-the-new-data-frontier"><b>Blog announcement</b></a>
+</p>
+
+**Monad** is a 56 million parameters generalist Small Reasoning Model, trained on 200 billions tokens from <a href="https://huggingface.co/PleIAs/Baguettotron">SYNTH</a>, a fully open generalist dataset.
+
+As of 2025, Monad is the best contender for the smallest viable language models. Despite being less than half of gpt-2, Monad not only answers in consistent English but performs significanly beyond chance on MMLU and other major industry benchmarks.
+
+<p align="center">
+  <img width="80%" src="figures/training_efficiency.jpeg">
+</p>
+
+Monad's name is a reference to Leibniz concept and general idea of the smallest possible unit of intelligence.
+
+## Features
+Monad has been natively trained for instructions with thinking traces. We implemented a series of dedicated pipelines for:
+* Memorization of encyclopedic knowledge (50,000 vital articles from Wikipedia), though in this size range hallucinations have to be expected.
+* Retrieval-Augmented Generation with grounding (following on our initial experiments with Pleias-RAG series)
+* Arithmetic and simple math resolution problem
+* Editing tasks
+* Information extraction
+* Creative writing, including unusual synthetic exercises like lipograms or layout poems.
+
+Monad is strictly monolingual in English. We trained a new custom tokenizer (likely one of the smallest tokenizer to date, less than 8,000 individual tokens), exclusively trained on SYNTH so that we maintain a relatively good compression ratio. 
+
+## Model design and training
+Monad is a 56M parameters decoders with a standard Qwen/Llama-like design, except for its extremely compact size and overall opiniated architecture for depth (with 64 layers)
+<p align="center">
+  <img width="80%" src="figures/monad_structure.png">
+</p>
+
+Monad was trained on 16 h100 from Jean Zay (compute plan n°A0191016886). Full pre-training took a bit less than 6 hours.
+
+## Evaluation
+Monad attains performance on MMLU significantly beyond chance with close to 30% of positive rate. We also find non-random results on gsm8k (8%) and HotPotQA (8%)
+
+To our knowledge, there is no model remotely close in this size range for evaluation comparison. Spiritually and practically, Monad remains unique.
+
+## Use and deployment
+Monad has been trained on the standard instruction style from Qwen.
+
+```xml
+<|im_start|>user
+Who are you?<|im_end|>
+<|im_start|>assistant
+<think>
+```
+
+Monad has no support yet for multi-turn.
+
+A major envisioned use case for Monad is explainability, as the model does provide a unique trade-off between observability and actual reasoning performance.

config.json

@@ -0,0 +1,31 @@
+{
+  "_attn_implementation_autoset": true,
+  "_name_or_path": "PleIAs/Monad",
+  "architectures": [
+    "LlamaForCausalLM"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "head_dim": 64,
+  "hidden_act": "silu",
+  "hidden_size": 256,
+  "initializer_range": 0.02,
+  "intermediate_size": 768,
+  "max_position_embeddings": 2048,
+  "mlp_bias": false,
+  "model_type": "llama",
+  "num_attention_heads": 4,
+  "num_hidden_layers": 64,
+  "num_key_value_heads": 4,
+  "pretraining_tp": 1,
+  "rms_norm_eps": 1e-05,
+  "rope_scaling": null,
+  "rope_theta": 10000,
+  "tie_word_embeddings": true,
+  "torch_dtype": "float32",
+  "transformers_version": "4.49.0",
+  "use_cache": true,
+  "vocab_size": 8192
+}

generation_config.json

@@ -0,0 +1,6 @@
+{
+  "_from_model_config": true,
+  "bos_token_id": 1,
+  "eos_token_id": 2,
+  "transformers_version": "4.49.0"
+}

onnx/model.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8243735cd3ca8f725f87aa5dee6142bb077a8a1db3a1f64087dab8554bf6e8c4
+size 227902592

onnx/model_bnb4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1bbb4bc65757e51e06f7831a75e00c8132e29f271de994262be19200869ebb85
+size 40539987

onnx/model_fp16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b9402c399e2d1be8de2249c30d6b6c11123cb05248367f9b64f371bcb4f1d38
+size 114619221

onnx/model_int8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9c27ae76f508e5aeb8b1954446831b9e7bde5d2da5cbf027cbb674bcdb2f0bd
+size 58526168

onnx/model_q4.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb6b89b38a81c43dd15860ea9b17760294e90ab7daac2771ef51f19f6cb8753e
+size 43944659

onnx/model_q4f16.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67000f2c517b89ed22ad425810fda68940962bb8ade11a26dd14d1de74e9516f
+size 36304936

onnx/model_quantized.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9c27ae76f508e5aeb8b1954446831b9e7bde5d2da5cbf027cbb674bcdb2f0bd
+size 58526168

onnx/model_uint8.onnx

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fec7db87faa8668021649d4899249b32ea6525c734ceda71a8f1ed6635b47737
+size 58526397

quantize_config.json

@@ -0,0 +1,18 @@
+{
+    "modes": [
+        "fp16",
+        "q8",
+        "int8",
+        "uint8",
+        "q4",
+        "q4f16",
+        "bnb4"
+    ],
+    "per_channel": false,
+    "reduce_range": false,
+    "block_size": null,
+    "is_symmetric": true,
+    "accuracy_level": null,
+    "quant_type": 1,
+    "op_block_list": null
+}

special_tokens_map.json

@@ -0,0 +1 @@
+{}

tokenizer_config.json

@@ -0,0 +1,40 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "[UNK]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<|begin_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "<|end_of_text|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "3": {
+      "content": "[PAD]",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "clean_up_tokenization_spaces": false,
+  "extra_special_tokens": {},
+  "model_max_length": 1000000000000000019884624838656,
+  "tokenizer_class": "PreTrainedTokenizer"
+}