| ---
|
| base_model: HuggingFaceTB/SmolLM2-1.7B-Instruct
|
| license: apache-2.0
|
| library_name: gguf
|
| tags:
|
| - gguf
|
| - llama.cpp
|
| - backpack
|
| ---
|
|
|
| # SmolLM2-1.7B-Instruct — Backpack GGUF
|
|
|
| 🎒 **Backpack Verified**
|
|
|
| GGUF quantizations of [HuggingFaceTB/SmolLM2-1.7B-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-1.7B-Instruct), tested for llama.cpp-compatible inference and packaged for Backpack.
|
|
|
| ## Model
|
|
|
| | Property | Value |
|
| | --- | --- |
|
| | Original model | `HuggingFaceTB/SmolLM2-1.7B-Instruct` |
|
| | Original publisher | HuggingFaceTB |
|
| | Upstream revision | `31b70e2e869a7173562077fd711b654946d38674` |
|
| | Architecture | LlamaForCausalLM |
|
| | Parameters | 1,711,376,384 |
|
| | Context length | 8,192 |
|
| | License | apache-2.0 |
|
|
|
| ## Available packages
|
|
|
| | Quantization | Size | Approx. RAM | Recommended for |
|
| | --- | ---: | ---: | --- |
|
| | Q4_K_M | 1006.7 MiB | 2.43 GB | Most users |
|
| | Q5_K_M | 1.1 GiB | 2.65 GB | Higher quality |
|
| | Q8_0 | 1.7 GiB | 3.46 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 SmolLM2-1.7B-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: 2026-08-22T16:19:59.672602+00:00
|
| - llama.cpp revision: `de699957b92f490efebad149665b0dccf127eaff`
|
| - SHA-256 checksums: see `checksums.sha256`
|
|
|
| - `SmolLM2-1.7B-Instruct-Q4_K_M.gguf`: `5ac93b9ab47d7b3069282f1cef65e1534e68b2c97d1ec129288bc792fa33ec5c`
|
| - `SmolLM2-1.7B-Instruct-Q5_K_M.gguf`: `a2fdd913de171cea397f3c876e9b23af9a899491e0dd32958affa2ec44c5d99c`
|
| - `SmolLM2-1.7B-Instruct-Q8_0.gguf`: `af66f269a1a77bd0b0df4f5a123b96582f1dac700b7846e0880b3eb54fa4f563`
|
|
|
|
|
| ## Provenance
|
|
|
| The source model was resolved to immutable revision `31b70e2e869a7173562077fd711b654946d38674`. 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/SmolLM2-1.7B-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.
|
|
|
