Update README.md

README.md

@@ -2,4 +2,212 @@
 base_model:
 - LiquidAI/LFM2-700M
 library_name: transformers.js
----
+license: other
+license_name: lfm1.0
+license_link: LICENSE
+language:
+  - en
+  - ar
+  - zh
+  - fr
+  - de
+  - ja
+  - ko
+  - es
+pipeline_tag: text-generation
+tags:
+  - liquid
+  - edge
+---
+
+
+<center>
+<div style="text-align: center;">
+  <img 
+    src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/7_6D7rWrLxp2hb6OHSV1p.png" 
+    alt="Liquid AI"
+    style="width: 100%; max-width: 66%; height: auto; display: inline-block; margin-bottom: 0.5em; margin-top: 0.5em;"
+  />
+</div>
+
+<a href="https://playground.liquid.ai/chat">
+<svg width="114.8" height="20" viewBox="0 0 1300 200" xmlns="http://www.w3.org/2000/svg" role="img" aria-label="Liquid Playground" style="margin-bottom: 1em;">
+  <title>Liquid: Playground</title>
+  <g>
+    <rect fill="#fff" width="600" height="200"></rect>
+    <rect fill="url(#x)" x="600" width="700" height="200"></rect>
+  </g>
+  <g transform="translate(20, 30) scale(0.4, 0.4)">
+    <path d="M172.314 129.313L172.219 129.367L206.125 188.18C210.671 195.154 213.324 203.457 213.324 212.382C213.324 220.834 210.956 228.739 206.839 235.479L275.924 213.178L167.853 33.6L141.827 76.9614L172.314 129.313Z" fill="black"/>
+    <path d="M114.217 302.4L168.492 257.003C168.447 257.003 168.397 257.003 168.352 257.003C143.515 257.003 123.385 237.027 123.385 212.387C123.385 203.487 126.023 195.204 130.55 188.24L162.621 132.503L135.966 86.7327L60.0762 213.183L114.127 302.4H114.217Z" fill="black"/>
+    <path d="M191.435 250.681C191.435 250.681 191.43 250.681 191.425 250.686L129.71 302.4H221.294L267.71 226.593L191.435 250.686V250.681Z" fill="black"/>
+  </g>
+  <g aria-hidden="true" fill="#fff" text-anchor="start" font-family="Verdana,DejaVu Sans,sans-serif" font-size="110">
+    <text x="200" y="148" textLength="329" fill="#000" opacity="0.1">Liquid</text>
+    <text x="190" y="138" textLength="329" fill="#000">Liquid</text>
+    <text x="655" y="148" textLength="619" fill="#000" opacity="0.1">Playground</text>
+    <text x="645" y="138" textLength="619">Playground</text>
+  </g>
+  
+  <linearGradient id="x" x1="0%" y1="0%" x2="100%" y2="0%">
+    <stop offset="0%" style="stop-color:#000000"></stop>
+    <stop offset="100%" style="stop-color:#000000"></stop>
+  </linearGradient>
+</svg>
+</a>
+</center>
+
+# LFM2-700M
+
+LFM2 is a new generation of hybrid models developed by [Liquid AI](https://www.liquid.ai/), specifically designed for edge AI and on-device deployment. It sets a new standard in terms of quality, speed, and memory efficiency. 
+
+We're releasing the weights of three post-trained checkpoints with 350M, 700M, and 1.2B parameters. They provide the following key features to create AI-powered edge applications:
+
+* **Fast training & inference** – LFM2 achieves 3x faster training compared to its previous generation. It also benefits from 2x faster decode and prefill speed on CPU compared to Qwen3.
+* **Best performance** – LFM2 outperforms similarly-sized models across multiple benchmark categories, including knowledge, mathematics, instruction following, and multilingual capabilities.
+* **New architecture** – LFM2 is a new hybrid Liquid model with multiplicative gates and short convolutions.
+* **Flexible deployment** – LFM2 runs efficiently on CPU, GPU, and NPU hardware for flexible deployment on smartphones, laptops, or vehicles.
+
+Find more information about LFM2 in our [blog post](https://www.liquid.ai/blog/liquid-foundation-models-v2-our-second-series-of-generative-ai-models).
+
+## 📄 Model details
+
+Due to their small size, **we recommend fine-tuning LFM2 models on narrow use cases** to maximize performance. 
+They are particularly suited for agentic tasks, data extraction, RAG, creative writing, and multi-turn conversations. 
+However, we do not recommend using them for tasks that are knowledge-intensive or require programming skills.
+
+| Property            | Value                         |
+| ------------------- | ----------------------------- |
+| **Parameters**      | 742,489,344                   |
+| **Layers**          | 16 (10 conv + 6 attn)         |
+| **Context length**  | 32,768 tokens                 |
+| **Vocabulary size** | 65,536                        |
+| **Precision**       | bfloat16                      |
+| **Training budget** | 10 trillion tokens            |
+| **License**         | LFM Open License v1.0         |
+
+**Supported languages**: English, Arabic, Chinese, French, German, Japanese, Korean, and Spanish.
+
+**Generation parameters**: We recommend the following parameters:
+* `temperature=0.3`
+* `min_p=0.15`
+* `repetition_penalty=1.05`
+
+**Architecture**: Hybrid model with multiplicative gates and short convolutions: 10 double-gated short-range LIV convolution blocks and 6 grouped query attention (GQA) blocks.
+
+**Pre-training mixture**: Approximately 75% English, 20% multilingual, and 5% code data sourced from the web and licensed materials.
+
+**Training approach**:
+* Knowledge distillation using [LFM1-7B](https://www.liquid.ai/blog/introducing-lfm-7b-setting-new-standards-for-efficient-language-models) as teacher model
+* Very large-scale SFT on 50% downstream tasks, 50% general domains
+* Custom DPO with length normalization and semi-online datasets
+* Iterative model merging
+
+## 🏃 How to run LFM2
+
+### Transformers.js
+
+If you haven't already, you can install the [Transformers.js](https://huggingface.co/docs/transformers.js) JavaScript library from [NPM](https://www.npmjs.com/package/@huggingface/transformers) using:
+```bash
+npm i @huggingface/transformers
+```
+
+You can then generate text as follows:
+```js
+import { pipeline, TextStreamer } from "@huggingface/transformers";
+
+// Create a text generation pipeline
+const generator = await pipeline(
+  "text-generation",
+  "onnx-community/LFM2-700M-ONNX",
+  { dtype: "q4" },
+);
+
+// Define the list of messages
+const messages = [
+  { role: "system", content: "You are a helpful assistant." },
+  { role: "user", content: "What is the capital of France?" },
+];
+
+// Generate a response
+const output = await generator(messages, {
+    max_new_tokens: 512,
+    do_sample: false,
+    streamer: new TextStreamer(generator.tokenizer, { skip_prompt: true, skip_special_tokens: true}),
+});
+console.log(output[0].generated_text.at(-1).content);
+// The capital of France is Paris.
+```
+
+### ONNXRuntime
+
+```py
+from transformers import AutoConfig, AutoTokenizer
+import onnxruntime
+import numpy as np
+from huggingface_hub import hf_hub_download
+
+# 1. Load config, processor, and model
+model_id = "onnx-community/LFM2-700M-ONNX"
+config = AutoConfig.from_pretrained(model_id)
+tokenizer = AutoTokenizer.from_pretrained(model_id)
+filename = "model.onnx" # Options: "model.onnx", "model_fp16.onnx", "model_q4.onnx", "model_q4f16.onnx"
+model_path = hf_hub_download(repo_id=model_id, filename=f"onnx/{filename}") # Download the graph
+hf_hub_download(repo_id=model_id, filename=f"onnx/{filename}_data") # Download the weights
+session = onnxruntime.InferenceSession(model_path)
+
+## Set config values
+num_key_value_heads = config.num_key_value_heads
+head_dim = config.hidden_size // config.num_attention_heads
+num_hidden_layers = config.num_hidden_layers
+eos_token_id = config.eos_token_id
+hidden_size = config.hidden_size
+conv_L_cache = config.conv_L_cache
+layer_types = config.layer_types
+
+# 2. Prepare inputs
+prompt = "What is C. elegans?"
+messages = [{"role": "user", "content": prompt}]
+inputs = tokenizer.apply_chat_template(messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="np")
+input_ids = inputs['input_ids']
+attention_mask = inputs['attention_mask']
+batch_size = input_ids.shape[0]
+position_ids = np.tile(np.arange(0, input_ids.shape[-1]), (batch_size, 1))
+past_cache_values = {}
+for i in range(num_hidden_layers):
+  if layer_types[i] == 'full_attention':
+    for kv in ('key', 'value'):
+      past_cache_values[f'past_key_values.{i}.{kv}'] = np.zeros([batch_size, num_key_value_heads, 0, head_dim], dtype=np.float32)
+  elif layer_types[i] == 'conv':
+    past_cache_values[f'past_conv.{i}'] = np.zeros([batch_size, hidden_size, conv_L_cache], dtype=np.float32)
+  else:
+    raise ValueError(f"Unsupported layer type: {layer_types[i]}")
+
+# 3. Generation loop
+max_new_tokens = 1024
+generated_tokens = np.array([[]], dtype=np.int64)
+for i in range(max_new_tokens):
+  logits, *present_cache_values = session.run(None, dict(
+      input_ids=input_ids,
+      attention_mask=attention_mask,
+      position_ids=position_ids,
+      **past_cache_values,
+  ))
+
+  ## Update values for next generation loop
+  input_ids = logits[:, -1].argmax(-1, keepdims=True)
+  attention_mask = np.concatenate([attention_mask, np.ones_like(input_ids, dtype=np.int64)], axis=-1)
+  position_ids = position_ids[:, -1:] + 1
+  for j, key in enumerate(past_cache_values):
+    past_cache_values[key] = present_cache_values[j]
+  generated_tokens = np.concatenate([generated_tokens, input_ids], axis=-1)
+  if (input_ids == eos_token_id).all():
+    break
+
+  ## (Optional) Streaming
+  print(tokenizer.decode(input_ids[0]), end='', flush=True)
+print()
+
+# 4. Output result
+print(tokenizer.batch_decode(generated_tokens, skip_special_tokens=True)[0])
+```