whatmodel.sh / SmolLM2 360M Instruct

SmolLM2 360M Instruct

+ contribute
Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

SmolLM2 360M Instruct is an open-weights language model from Hugging Face, released October 2024. Its published size is 0.362B parameters and a 8k-token context window. Licence: apache-2.0.

Catalogue identity notes

HuggingFaceTB/SmolLM2-360M-Instruct @a10cc15 (created 2024-10-31, licence apache-2.0). Hugging Face SmolLM2 360M instruct model, sibling of the catalogued smollm2-1-7b. Lab rows from its own card's lighteval table with the catalogued Qwen2.5-0.5B-Instruct comparator; independent EuroEval row.

Model information sources

Release date ↗ · Weights licence ↗

HF safetensors metadata 361,821,120 parameters (HuggingFaceTB/SmolLM2-360M-Instruct @a10cc15).

18.6±2.7#693 of 719 LLMsfrom 6 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of SmolLM2 360M Instruct, out of 828 with a score, then more from Hugging Face.

#modelscorewhereRAMreleased
797Falcon-RW 1B TII20±2local~4.3GB2023-04
798OPT-2.7B Meta19±1local~7.1GB2022-05
799SmolLM2 360M Instruct Hugging Face19±3local~1.7GB2024-10
800GPT-Neo 2.7B EleutherAI18±1local~13GB2021-03
801GPT-3 babbage (API base model) OpenAI18±1comparison only2020-06
more from Hugging Face
451StarChat2 15B v0.1 Hugging Face48±1local~12GB2024-03
532Zephyr ORPO 141B-A35B v0.1 Hugging Face45±2local~56GB2024-04
577SmolLM3 3B Hugging Face43±1local~3.3GB2025-07

see all LLMs · find the ones that fit your machine

$ what it's good at

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #342 of 393 * · 9.5% of its score · 1 test

Beats Gemma 4 E2B, Code Llama 70B Instruct, QwQ-32B-Preview here, though they rank higher overall. Trails Gemma 3 270M IT here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
IFEval (aggregation unspecified) lab41.0% ✎▁▃▃▁▂▄▄▆█▃▂▁ beats 2/141easy · avg 77%▇▇▇▇ 9.5%at its level · small run
math█··········█ #2 of 2 * · 30% of its score · 1 test
testresultvs the others in its rundifficultycountswhy
GSM8K lab7.4% ✎▃▂▂▅▅▆▅▇█▇▄▃ beats 0/146easy · avg 83%▇▇▇▇▇▇▇▇▇▇▇▇ 30%sharp · small run
other tests4 not in a task · 57% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard lab27.3% ✎▁▁▃▁▂▄▅▅█▃▄▁ beats 2/129medium · avg 69%▇▇▇▇▇▇▇▇▇▇▇▇▇ 33%sharp · too hard for it · small run
HellaSwag lab52.1% ✎▁▁▁▁▂▃▃▇█▆▅▃ beats 13/85easy · avg 87%▇▇▇▇▇▇ 15%at its level · small run
PIQA lab70.8% ✎▁▁▁▃▃▄▇█▄▃▁▁ beats 15/83easy · avg 86%▇▇▇ 6.8%blunt · at its level · small run
EuroEval HellaSwag (English), test split, 5-shot generative MC, accuracy24.9% ✎▃·▁▁▂▁▂▅█▇▄▁ beats 6/100medium · avg 68%▇ 2.3%too hard for it · noisy

$ where to run it

llama.cpp · recommended

  1. Install llama.cpp · Homebrew and winget installs name the commands llama-cli and llama-server instead.
    macOS / Linux
    curl -LsSf https://llama.app/install.sh | sh
    Windows
    irm https://llama.app/install.ps1 | iex

    other ways to install ↗

  2. Run SmolLM2 360M Instruct · needs ~0.39 GB · from huggingface.co/HuggingFaceTB
    chat in the terminal
    llama cli -hf HuggingFaceTB/SmolLM2-360M-Instruct-GGUF:Q8_0
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf HuggingFaceTB/SmolLM2-360M-Instruct-GGUF:Q8_0

LM Studio

version

  1. Install LM Studio · Open the app once before using the lms command.

    download it (macOS / Windows / Linux) ↗

  2. Run SmolLM2 360M Instruct · needs ~0.2 GB, Mac only · from huggingface.co/Irfanuruchi

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/Irfanuruchi/SmolLM2-360M-Instruct-MLX-4bit -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run SmolLM2 360M Instruct · needs ~0.726 GB · from ollama.com
    chat in the terminal
    ollama run smollm2:360m

MLX (Mac)

version

  1. Install MLX (Mac) · Apple silicon Macs only, text only.
    with uv
    uv tool install mlx-lm
    or
    pip install mlx-lm
  2. Run SmolLM2 360M Instruct · needs ~0.2 GB, Mac only · from huggingface.co/Irfanuruchi
    chat in the terminal
    mlx_lm.chat --model Irfanuruchi/SmolLM2-360M-Instruct-MLX-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model Irfanuruchi/SmolLM2-360M-Instruct-MLX-4bit

Weights and documentation: huggingface.co/HuggingFaceTB ↗

$ how the score was worked out · HuggingFaceTB/SmolLM2-360M-In… 94%, EuroEval leaderboards (result… 2%

Each result implies a score on its own, after adjusting for how hard its test is and who ran it. The score is their weighted average, with a small pull toward the middle for thinly tested models. Each test's weight is its counts column above; here it's added up by who ran the tests. How it works

sourceresultsshare of the scorelicence
HuggingFaceTB/SmolLM2-360M-Instruct model card · lab-reported5▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 94%Apache-2.0
EuroEval leaderboards (results collection), English generative leaderboard1▇ 2%MIT
prior · pull toward the middle▇ 3%
weighted average= 18.6

$ improve this page

Update model informationAdd a place to run itAdd a test resultPropose a benchmarkCorrect a benchmark scoreEdit data on GitHub

Use a short GitHub form or edit the data through a pull request. Every number needs a source; here's which ones we can use.

data updated · sources and licences · json