whatmodel.sh / Qwen2.5 0.5B Instruct

Qwen2.5 0.5B Instruct

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

Qwen2.5 0.5B Instruct is an open-weights language model from Alibaba, released September 2024. Its published size is 0.49B parameters and a 33k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo Qwen/Qwen2.5-0.5B-Instruct @7ae5576: licence tag apache-2.0. 494,032,768 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-09: month the maker's Hugging Face repo was created (createdAt 2024-09-16). Ollama tags qwen2.5:0.5b-instruct-q4_K_M, qwen2.5:0.5b-instruct-q8_0 are the library's instruct/chat builds of this checkpoint (tag names on the pinned tags page); download sizes are estimated from the parameter count.

26.5±1.8#666 of 719 LLMsfrom 12 benchmark results across 7 sources

$ how it compares

The LLMs ranked either side of Qwen2.5 0.5B Instruct, out of 828 with a score, then more from Alibaba.

#modelscorewhereRAMreleased
766Gemma 2B IT Google27±1local~3.1GB2024-02
767EuroLLM 1.7B Instruct UTTER (EuroLLM)27±3local~2.4GB2024-08
768Qwen2.5 0.5B Instruct Alibaba27±2local~1.8GB2024-09
769GPT-3 text-davinci-001 OpenAI26±3comparison only2022-01
770GPT4All-13B-Snoozy Nomic AI26±2local~56GB2023-04
more from Alibaba
18Qwen3.8 Max (2.4T-A95B) Alibaba78±1api $3.00/M +local~380GB2026-08
34Qwen3.8 Max (0902) Alibaba76±1api $3.00/M2026-09
37Qwen3.8-Flash-Next Alibaba75±1local~59GB2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #290 of 300 * · 5.0% of its score · 1 test

Beats QwQ-32B-Preview, DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B), Phi-4-mini 3.8B here, though they rank higher overall. Trails Granite 4.0 350M here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
BFCL v335.6% ✎▁▁▁▁▃▃▅█▅▄▁▁ beats 10/114medium · avg 52%▇▇ 5.0%blunt · at its level
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #354 of 393 * · 9.3% of its score · 2 tests

Beats Gemma 4 E2B, Code Llama 70B Instruct, Phi-3-mini-4k-instruct (June 2024 update) here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct here, though they rank lower overall.

testresultvs the others in its rundifficultycountswhy
Arabic IFEval prompt-level (Inception)13.9% ✎▁·▁▁▂▁▄▄▆▄█▄ beats 7/38medium · avg 49%▇▇ 5.0%sharp · too hard for it
IFEval (aggregation unspecified)31.6% ✎▃▁▃▁▂▄▄▆█▃▂▁ beats 0/141easy · avg 77%▇▇ 4.3%at its level · small run
math█··········█ #1 of 2 * · 16% of its score · 1 test
testresultvs the others in its rundifficultycountswhy
GSM8K26.8% ✎▂▂▂▅▅▆▅▇█▇▄▃ beats 17/146easy · avg 83%▇▇▇▇▇▇▇ 16%sharp · small run
other tests8 not in a task · 68% of its score
testresultvs the others in its rundifficultycountswhy
BIG-Bench Hard30.7% ✎▁▁▁▁▃▄▅▅█▃▄▁ beats 5/129medium · avg 69%▇▇▇▇▇▇▇▇ 20%sharp · too hard for it · small run
CPTUB average (SpeakLeash, 0-5 judge scale)28.0% ✎▃▂▂▂▃█▃▆▅▄▂▃ beats 0/95easy · avg 66%▇▇▇▇▇▇▇ 18%at its level
Open Persian LLM Leaderboard v2 average (21 tasks)33.3% ✎▂▄▂▃█▄▃▆▃▅▂▂ beats 4/81medium · avg 58%▇▇▇▇ 10%blunt · at its level
HellaSwag48.0% ✎▁▁▁▁▂▃▃▇█▆▅▃ beats 10/85easy · avg 87%▇▇▇ 6.7%at its level · small run
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)74.8% ✎▃▁▂▃▄▄█▆▃▂▂▁ beats 2/140very easy · avg 94%▇▇ 4.1%blunt · noisy
Open PT-LLM Leaderboard average (9 Portuguese tasks)50.9% ✎▃▁▁▁▁▁▁▃▂█▅▁ beats 71/276medium · avg 73%▇ 3.7%noisy
PIQA67.2% ✎▁▁▁▃▃▄▇█▄▃▁▁ beats 9/83easy · avg 86%▇ 3.0%blunt · at its level · small run
AraGen-v3 3C3H score (Inception, Claude Sonnet 4 judge)5.2% ✎▁▁·▁▃▂▄█▄▅▄▄ beats 4/66hard · avg 31%▇ 2.7%too hard for it

$ where to run it

llama.cpp · recommended

version

  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 Qwen2.5 0.5B Instruct · needs ~0.5 GB · from huggingface.co/Qwen (official)
    chat in the terminal
    llama cli -hf Qwen/Qwen2.5-0.5B-Instruct-GGUF:q4_k_m
  3. Or start a local server and open the address it prints (http://localhost:9931)
    llama serve -hf Qwen/Qwen2.5-0.5B-Instruct-GGUF:q4_k_m

LM Studio

version

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

    download it (macOS / Windows / Linux) ↗

  2. Run Qwen2.5 0.5B Instruct · needs ~0.5 GB · from huggingface.co/Qwen (official)

    open in LM Studio ↗

  3. Or download it in the terminal
    lms get https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF@q4_k_m -y
  4. Then run lms ls to see its name, and lms chat followed by that name

Ollama

version

  1. Install Ollama

    download it (macOS / Windows) ↗

    Linux
    curl -fsSL https://ollama.com/install.sh | sh
  2. Run Qwen2.5 0.5B Instruct · needs ~0.79 GB · from ollama.com
    chat in the terminal
    ollama run qwen2.5:0.5b-instruct-q4_K_M

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 Qwen2.5 0.5B Instruct · needs ~0.3 GB, Mac only · from huggingface.co/lmstudio-community
    chat in the terminal
    mlx_lm.chat --model lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-4bit
  3. Or start an OpenAI-compatible server at http://localhost:8080/v1
    mlx_lm.server --model lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · HuggingFaceTB/SmolLM2-360M-In… 50%, SpeakLeash CPTUB (Complex Pol… 18%, Open Persian LLM Leaderboard… 10% and 4 more

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 card5▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 50%Apache-2.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)1▇▇▇▇▇ 18%MIT
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)1▇▇▇ 10%Apache-2.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)2▇▇ 8%Apache-2.0
Berkeley Function Calling Leaderboard result archive1▇▇ 5%Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)1▇ 4%Apache-2.0
Open Portuguese LLM Leaderboard (archived)1▇ 4%Apache-2.0
prior · pull toward the middle▇ 1%
weighted average= 26.5

$ 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