Add files using upload-large-folder tool

.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
+SmolLM2-135M-Instruct-Q8_0.gguf filter=lfs diff=lfs merge=lfs -text
+SmolLM2-135M-Instruct-Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+SmolLM2-135M-Instruct-Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,86 @@
+---
+base_model: HuggingFaceTB/SmolLM2-135M-Instruct
+license: apache-2.0
+library_name: gguf
+tags:
+  - gguf
+  - llama.cpp
+  - backpack
+---
+
+# SmolLM2-135M-Instruct — Backpack GGUF
+
+🎒 **Backpack Verified**
+
+GGUF quantizations of [HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct), tested for llama.cpp-compatible inference and packaged for Backpack.
+
+## Model
+
+| Property | Value |
+| --- | --- |
+| Original model | `HuggingFaceTB/SmolLM2-135M-Instruct` |
+| Original publisher | HuggingFaceTB |
+| Upstream revision | `12fd25f77366fa6b3b4b768ec3050bf629380bac` |
+| Architecture | LlamaForCausalLM |
+| Parameters | 134,515,008 |
+| Context length | 8,192 |
+| License | apache-2.0 |
+
+## Available packages
+
+| Quantization | Size | Approx. RAM | Recommended for |
+| --- | ---: | ---: | --- |
+| Q4_K_M | 100.6 MiB | 1.14 GB | Most users |
+| Q5_K_M | 106.9 MiB | 1.15 GB | Higher quality |
+| Q8_0 | 138.1 MiB | 1.2 GB | Plenty of memory |
+
+
+Memory values are estimates, not guarantees. Runtime configuration and context length change actual use.
+
+## Backpack recommendation
+
+**Recommended: Q4_K_M.** It usually offers a practical quality, size, and speed balance for local inference.
+
+## Run with llama.cpp
+
+Using the llama.cpp revision recorded below:
+
+```bash
+llama-cli --model SmolLM2-135M-Instruct-Q4_K_M.gguf --conversation
+```
+
+## Run with Backpack
+
+These artifacts and `backpack-model.yaml` are prepared for the [Backpack AI workspace](https://backpack.run).
+
+## Validation
+
+| Package | Integrity | Load | Inference | Tokenizer |
+| --- | --- | --- | --- | --- |
+| Q4_K_M | passed | passed | passed | passed |
+| Q5_K_M | passed | passed | passed | passed |
+| Q8_0 | passed | passed | passed | passed |
+
+
+- Packaged: 2026-08-20T20:32:10.805207+00:00
+- llama.cpp revision: `de699957b92f490efebad149665b0dccf127eaff`
+- SHA-256 checksums: see `checksums.sha256`
+
+- `SmolLM2-135M-Instruct-Q4_K_M.gguf`: `dd18a11b8634d1684448986b8c166f75319f52082d759654aaa8fe5bd2f057e3`
+- `SmolLM2-135M-Instruct-Q5_K_M.gguf`: `00680963c363ba10593daf7568dd6e1ee4c4771a608fe1b4e43f86b564d9b823`
+- `SmolLM2-135M-Instruct-Q8_0.gguf`: `ee785d9b4836ddb57207ae6daa630206a756c99fc52e19696f1e2ea2e8a41b99`
+
+
+## Provenance
+
+The source model was resolved to immutable revision `12fd25f77366fa6b3b4b768ec3050bf629380bac`. It was converted with llama.cpp's `convert_hf_to_gguf.py` and quantized with `llama-quantize`; the exact tested revision is recorded above and in `backpack-model.yaml`.
+
+## License and attribution
+
+Upstream declares **apache-2.0**. Review the [upstream model card](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct) and comply with all applicable terms.
+
+Backpack does not claim ownership of the original model. These artifacts are packaged and quantized distributions of the upstream model.
+
+## Disclaimer
+
+Quantization can alter output quality. Memory estimates vary with runtime configuration, context length, and hardware.

SmolLM2-135M-Instruct-Q4_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd18a11b8634d1684448986b8c166f75319f52082d759654aaa8fe5bd2f057e3
+size 105453984

SmolLM2-135M-Instruct-Q5_K_M.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:00680963c363ba10593daf7568dd6e1ee4c4771a608fe1b4e43f86b564d9b823
+size 112103328

SmolLM2-135M-Instruct-Q8_0.gguf

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee785d9b4836ddb57207ae6daa630206a756c99fc52e19696f1e2ea2e8a41b99
+size 144810912

backpack-model.yaml

@@ -0,0 +1,106 @@
+schema_version: 1
+model:
+  id: smollm2-135m-instruct
+  display_name: SmolLM2-135M-Instruct
+  architecture: LlamaForCausalLM
+  model_type: llama
+  parameter_count: 134515008
+  context_length: 8192
+  tokenizer_type: GPT2Tokenizer
+  tasks:
+  - text-generation
+  chat_template: true
+  multimodal: false
+  custom_code: false
+  safetensors: true
+upstream:
+  repo: HuggingFaceTB/SmolLM2-135M-Instruct
+  revision: 12fd25f77366fa6b3b4b768ec3050bf629380bac
+  license:
+    identifier: apache-2.0
+    redistribution: allowed
+    reason: Upstream declares a recognized license that permits redistribution; retain
+      its terms and attribution.
+  gated: false
+  private: false
+  source_url: https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct
+packages:
+- id: gguf-q4-k-m
+  format: gguf
+  precision: Q4_K_M
+  filename: SmolLM2-135M-Instruct-Q4_K_M.gguf
+  sha256: dd18a11b8634d1684448986b8c166f75319f52082d759654aaa8fe5bd2f057e3
+  size_bytes: 105453984
+  bits_per_weight: 4.83
+  runtime:
+    provider: llama.cpp
+    tested_revision: de699957b92f490efebad149665b0dccf127eaff
+  hardware:
+    estimated_ram_gb: 0.13
+    estimated_vram_gb: 0.12
+    recommended_ram_gb: 1.14
+    note: Estimate; runtime use varies with context length and configuration.
+  validation:
+    integrity: passed
+    metadata: passed
+    load: passed
+    inference: passed
+    tokenizer: passed
+    tested_at: '2026-08-20T20:24:16.685253Z'
+    details:
+      version: '3'
+      tensor_count: '272'
+      metadata_count: '30'
+- id: gguf-q5-k-m
+  format: gguf
+  precision: Q5_K_M
+  filename: SmolLM2-135M-Instruct-Q5_K_M.gguf
+  sha256: 00680963c363ba10593daf7568dd6e1ee4c4771a608fe1b4e43f86b564d9b823
+  size_bytes: 112103328
+  bits_per_weight: 5.67
+  runtime:
+    provider: llama.cpp
+    tested_revision: de699957b92f490efebad149665b0dccf127eaff
+  hardware:
+    estimated_ram_gb: 0.13
+    estimated_vram_gb: 0.12
+    recommended_ram_gb: 1.15
+    note: Estimate; runtime use varies with context length and configuration.
+  validation:
+    integrity: passed
+    metadata: passed
+    load: passed
+    inference: passed
+    tokenizer: passed
+    tested_at: '2026-08-20T20:24:18.938307Z'
+    details:
+      version: '3'
+      tensor_count: '272'
+      metadata_count: '30'
+- id: gguf-q8-0
+  format: gguf
+  precision: Q8_0
+  filename: SmolLM2-135M-Instruct-Q8_0.gguf
+  sha256: ee785d9b4836ddb57207ae6daa630206a756c99fc52e19696f1e2ea2e8a41b99
+  size_bytes: 144810912
+  bits_per_weight: 8.5
+  runtime:
+    provider: llama.cpp
+    tested_revision: de699957b92f490efebad149665b0dccf127eaff
+  hardware:
+    estimated_ram_gb: 0.17
+    estimated_vram_gb: 0.16
+    recommended_ram_gb: 1.2
+    note: Estimate; runtime use varies with context length and configuration.
+  validation:
+    integrity: passed
+    metadata: passed
+    load: passed
+    inference: passed
+    tokenizer: passed
+    tested_at: '2026-08-20T20:24:21.225025Z'
+    details:
+      version: '3'
+      tensor_count: '272'
+      metadata_count: '30'
+generated_at: '2026-08-20T20:32:10.805207Z'

checksums.sha256

@@ -0,0 +1,3 @@
+dd18a11b8634d1684448986b8c166f75319f52082d759654aaa8fe5bd2f057e3  SmolLM2-135M-Instruct-Q4_K_M.gguf
+00680963c363ba10593daf7568dd6e1ee4c4771a608fe1b4e43f86b564d9b823  SmolLM2-135M-Instruct-Q5_K_M.gguf
+ee785d9b4836ddb57207ae6daa630206a756c99fc52e19696f1e2ea2e8a41b99  SmolLM2-135M-Instruct-Q8_0.gguf