Add SmolLM2 135M Instruct F16 and Q6_K GGUF artifacts

.gitattributes

@@ -33,3 +33,5 @@ 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
+SmolLM2-135M-Instruct-Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
+SmolLM2-135M-Instruct-f16.gguf filter=lfs diff=lfs merge=lfs -text

CHECKSUMS.sha256

@@ -0,0 +1,2 @@
+fd659c45cacd2213d6b8ffe002c261c193eed4cc0d2b1b2af836fba7b030177c  SmolLM2-135M-Instruct-f16.gguf
+e12aa5cace7cca9c6bd27a8eddb511d5140628ca5fe3f3f3d20fbbf5c458aeee  SmolLM2-135M-Instruct-Q6_K.gguf

README.md

@@ -0,0 +1,63 @@
+---
+language:
+- en
+license: apache-2.0
+library_name: llama.cpp
+pipeline_tag: text-generation
+tags:
+- gguf
+- llama.cpp
+- smollm2
+- text-generation
+- conversational
+base_model: HuggingFaceTB/SmolLM2-135M-Instruct
+---
+
+# SmolLM2 135M Instruct — GGUF
+
+This repository provides local CPU-oriented GGUF conversions of [`HuggingFaceTB/SmolLM2-135M-Instruct`](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct) for use with `llama.cpp` and compatible runtimes. The original instruct model is a compact 135M-parameter model released under the Apache-2.0 license.
+
+## Available files
+
+| File | Format | Intended use | File size |
+|---|---|---|---:|
+| `SmolLM2-135M-Instruct-f16.gguf` | F16 GGUF | Quality-preserving local inference | 258 MiB |
+| `SmolLM2-135M-Instruct-Q6_K.gguf` | Q6_K GGUF | Lower-memory CPU inference | 132 MiB |
+
+The F16 variant preserves the original converted weight precision. The Q6_K variant is provided for systems where memory usage is more important than retaining the F16 representation. Use the F16 file for the highest-fidelity local behavior.
+
+## Quick start with llama.cpp
+
+```bash
+./llama-cli \
+  --model SmolLM2-135M-Instruct-f16.gguf \
+  --conversation \
+  --n-gpu-layers 0
+```
+
+For persistent local serving, start `llama-server` once and send OpenAI-compatible requests to the local endpoint:
+
+```bash
+./llama-server \
+  --model SmolLM2-135M-Instruct-f16.gguf \
+  --host 127.0.0.1 \
+  --port 8080 \
+  --ctx-size 2048 \
+  --n-gpu-layers 0
+```
+
+## Integrity verification
+
+Verify downloaded files with:
+
+```bash
+sha256sum -c CHECKSUMS.sha256
+```
+
+## Conversion details
+
+The model was converted from the official Hugging Face checkpoint with the official `llama.cpp` Hugging Face-to-GGUF converter. The F16 GGUF output was retained as the quality-preserving variant, then quantized with `llama-quantize` to produce the Q6_K variant. See `conversion-metadata.json` for the artifact metadata.
+
+## License and attribution
+
+These GGUF artifacts are derived from [HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct). The upstream model card identifies the model license as **Apache-2.0**. Retain upstream attribution and consult the original model card for limitations, training details, and citation information.

SmolLM2-135M-Instruct-Q6_K.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e12aa5cace7cca9c6bd27a8eddb511d5140628ca5fe3f3f3d20fbbf5c458aeee
+size 138382912

SmolLM2-135M-Instruct-f16.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd659c45cacd2213d6b8ffe002c261c193eed4cc0d2b1b2af836fba7b030177c
+size 270885952

conversion-metadata.json

@@ -0,0 +1,22 @@
+{
+  "base_model": "HuggingFaceTB/SmolLM2-135M-Instruct",
+  "base_model_url": "https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct",
+  "license": "Apache-2.0",
+  "converter": "llama.cpp convert_hf_to_gguf.py",
+  "quantizer": "llama.cpp llama-quantize",
+  "artifacts": [
+    {
+      "file": "SmolLM2-135M-Instruct-f16.gguf",
+      "quantization": "F16",
+      "bytes": 270885952,
+      "sha256": "fd659c45cacd2213d6b8ffe002c261c193eed4cc0d2b1b2af836fba7b030177c"
+    },
+    {
+      "file": "SmolLM2-135M-Instruct-Q6_K.gguf",
+      "quantization": "Q6_K",
+      "bytes": 138382912,
+      "sha256": "e12aa5cace7cca9c6bd27a8eddb511d5140628ca5fe3f3f3d20fbbf5c458aeee"
+    }
+  ]
+}
+