GGUF release: F16/Q8_0/Q4_0 + card aligned with the encoder release card

.gitattributes

@@ -33,3 +33,6 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+LFM2.5-Encoder-230M-F16.gguf filter=lfs diff=lfs merge=lfs -text
+LFM2.5-Encoder-230M-Q4_0.gguf filter=lfs diff=lfs merge=lfs -text
+LFM2.5-Encoder-230M-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text

LFM2.5-Encoder-230M-F16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1516abfdb4ac0befcad6aa547204e1d26e8b9c83da46891a5fc289a6280e5c53
+size 461882336

LFM2.5-Encoder-230M-Q4_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:476b1347517eee013337ad77779d2ee32c4df308f19bccef83379d040b66c868
+size 149079008

LFM2.5-Encoder-230M-Q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0e92be849493abd5c0ff356416fffadf42cb52cacc717181e75f6be6ba1838e4
+size 246596576

LICENSE

@@ -1,4 +1,4 @@
-LFM Open License v1.0
+LFM Open License v1.0 
 
 TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
 
@@ -68,4 +68,4 @@ You may add Your own copyright statement to Your modifications and may provide a
 
 11. Termination. This License will terminate automatically and immediately if You fail to comply with any of its terms and conditions. Upon termination, You must cease all use of the Work and any Derivative Works and delete all copies in Your possession.
 
-END OF TERMS AND CONDITIONS
+END OF TERMS AND CONDITIONS

README.md

@@ -1,52 +1,80 @@
 ---
-library_name: gguf
-license: other
-license_name: lfm1.0
-license_link: LICENSE
 language:
 - en
-- ar
-- zh
-- fr
 - de
-- ja
-- ko
 - es
-- pt
+- fr
 - it
-pipeline_tag: text-generation
-base_model: LiquidAI/LFM2.5-Encoder-230M
+- nl
+- pl
+- pt
+- ar
+- hi
+- ja
+- ru
+- tr
+- vi
+- zh
 tags:
 - liquid
+- lfm2
 - lfm2.5
-- gguf
+- bidirectional
+- masked-lm
+- encoder
 - llama.cpp
+- gguf
+pipeline_tag: fill-mask
+license: other
+license_name: lfm1.0
+license_link: LICENSE
+base_model:
+- LiquidAI/LFM2.5-Encoder-230M
 ---
 
-<div align="center">
-  <img
-    src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png"
+<center>
+<div style="text-align: center;">
+  <img 
+    src="https://cdn-uploads.huggingface.co/production/uploads/61b8e2ba285851687028d395/2b08LKpev0DNEk6DlnWkY.png" 
     alt="Liquid AI"
     style="width: 100%; max-width: 100%; height: auto; display: inline-block; margin-bottom: 0.5em; margin-top: 0.5em;"
   />
-  <div style="display: flex; justify-content: center; gap: 0.5em; margin-bottom: 1em;">
-    <a href="https://playground.liquid.ai/"><strong>Try LFM</strong></a> •
-    <a href="https://docs.liquid.ai/lfm/getting-started/welcome"><strong>Docs</strong></a> •
-    <a href="https://leap.liquid.ai/"><strong>LEAP</strong></a> •
-    <a href="https://discord.com/invite/liquid-ai"><strong>Discord</strong></a>
-  </div>
 </div>
+<div style="display: flex; justify-content: center; gap: 0.5em;">
+<a href="https://playground.liquid.ai/"><strong>Try LFM</strong></a> • <a href="https://docs.liquid.ai/lfm"><strong>Documentation</strong></a> • <a href="https://leap.liquid.ai/"><strong>LEAP</strong></a>
+</div>
+</center>
 
-# LFM2.5-Encoder-230M-GGUF
+# LFM2.5-Encoder-230M
 
-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.
+LFM2.5-Encoder-230M is a multilingual bidirectional encoder built on the LFM2 architecture — a lightweight encoder for tight latency and memory budgets, punching above its size class. It is a masked language model with full bidirectional attention, designed to be fine-tuned into task-specific models (classification, token classification, retrieval, reranking, and semantic similarity) across 15 languages, and to run efficiently on-device.
 
-Find more details in the original model card: https://huggingface.co/LiquidAI/LFM2.5-Encoder-230M
+- **Highly capable for its size.** On par with the best similarly sized encoders and well ahead of our own retrieval siblings.
+- **General-purpose.** 8k context, strong across NLI, paraphrase, sentiment, and multilingual tasks.
+- **Fast and on-device.** Matches or beats ModernBERT throughput, with a long-context edge on CPU; runs in the browser on WebGPU.
 
-## 🏃 How to run LFM2
+Find more information about LFM2.5-Encoder-230M in our [blog post](https://www.liquid.ai/blog/lfm2-5-encoders).
+
+## 🏃 How to run
 
 Example usage with [llama.cpp](https://github.com/ggml-org/llama.cpp):
 
+Run masked-token prediction — the prompt's literal `[MASK]` is replaced with the model's mask token, one non-causal forward pass is run, and the top-K predictions at the mask position are printed:
+
+```bash
+llama-fill-mask -hf LiquidAI/LFM2.5-Encoder-230M-GGUF "The capital of France is [MASK]." 5
+#   1    16.17  ' Paris'
+#   2    13.41  ' Strasbourg'
+#   3    13.35  'Paris'
+#   4    13.18  ' Lyon'
+#   5    11.87  ' Versailles' 
 ```
-llama-cli -hf LiquidAI/LFM2.5-Encoder-230M-GGUF
+
+For downstream use, the encoder body also serves per-token embeddings:
+
+```bash
+llama-server -hf LiquidAI/LFM2.5-Encoder-230M-GGUF --embeddings
+curl -s http://localhost:8080/embedding -d '{"content": "hello world"}'
 ```
+
+Find more details in the original model card: https://huggingface.co/LiquidAI/LFM2.5-Encoder-230M