[whatmodel.sh](../../README.md) / Qwen2 0.5B Instruct
# Qwen2 0.5B Instruct
Alibaba · released 2024-06 · 0.49B params · apache-2.0
**score 27.9** ±1.6 · #677 of 732 · from 8 benchmark results across 4 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks Qwen2 0.5B Instruct among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ●··········█ #2 of 2* 14% 1
chat & writing ●··········█ #2 of 2* 4.8% 1
math ▁▁●▅█▆▄█▄▂▁▁ #411 of 431* 17% 1
coding · #2 of 2 · 14% of its score
test result vs the others in its run difficulty counts why
HumanEval (maker) 17.1% ▁▁▁▁▂●▆▆█▅▂▁ beats 29/205 easy · avg 80% 14% sharp · small run
chat & writing · #2 of 2 · 4.8% of its score
test result vs the others in its run difficulty counts why
IFEval prompt-level strict (maker) 20.0% ●▁··▂··▂▇█▃▃ beats 0/26 medium · avg 64% 4.8% at its level · small run
math · #411 of 431 · 17% of its score
Beats [Command R (03-2024)](command-r-03-2024.md), [Llama 2 13B Chat](llama-2-13b-chat.md), [Qwen2 1.5B Instruct](qwen2-1-5b.md) here, though they rank higher overall. Trails [GPT-Neo 125M](gpt-neo-125m.md), [OPT-125M](opt-125m.md), [GPT-2 XL (1.5B)](gpt-2-xl.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K (maker) 40.1% ▂▂▂▂▃▄●▇██▆▄ beats 73/206 very easy · avg 92% 17% sharp · at its level · small run
other tests (5)
test result vs the others in its run difficulty counts why
LiveBench 2024-07-26 7.7% ▁●▁▂▂▃▂█▆▃▅▃ beats 2/61 medium · avg 38% 32% too hard for it
MMLU (maker) 37.9% ▁▁▁▁▁▂●▅██▆▄ beats 28/214 easy · avg 80% 10% small run
Open Persian LLM Leaderboard v2 a… 30.6% ●▄▂▃█▄▃▆▃▅▂▂ beats 1/81 medium · avg 57% 10% blunt · at its level
C-Eval (maker) 45.2% ▁···▁▁●▁▂▂▃█ beats 5/33 easy · avg 81% 7.1% at its level · small run
Open PT-LLM Leaderboard average (… 44.4% ▃▁▁▁▁▁●▁▂█▅▁ beats 57/276 medium · avg 72% 3.6% noisy
source results share of the score
Qwen2-0.5B-Instruct model card (submitt… 5 53% ▇▇▇▇▇▇
LiveBench model judgments 1 32% ▇▇▇▇
Open Persian LLM Leaderboard v2 1 10% ▇
Open Portuguese LLM Leaderboard 1 4% ▇
prior: pull toward the middle 1% ▇
weighted average = 27.9
Licences: Qwen2-0.5B-Instruct model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · LiveBench model judgments ([Apache-2.0 (stated in the LiveBench datasheet)](https://www.apache.org/licenses/LICENSE-2.0)) · Open Persian LLM Leaderboard v2 ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run Qwen2 0.5B Instruct (Q4_K_M, needs ~0.4 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct-GGUF/tree/198f08841147e5196a6a69bd0053690fb1fd3857))
```sh
llama cli -hf Qwen/Qwen2-0.5B-Instruct-GGUF:q4_k_m
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf Qwen/Qwen2-0.5B-Instruct-GGUF:q4_k_m
```
Other versions: Q8_0 (~0.5 GB) `llama cli -hf Qwen/Qwen2-0.5B-Instruct-GGUF:q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run Qwen2 0.5B Instruct (Q4_K_M, needs ~0.4 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct-GGUF/tree/198f08841147e5196a6a69bd0053690fb1fd3857))
```sh
lms get https://huggingface.co/Qwen/Qwen2-0.5B-Instruct-GGUF@q4_k_m -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~0.5 GB) `lms get https://huggingface.co/Qwen/Qwen2-0.5B-Instruct-GGUF@q8_0 -y` · 4bit (~0.3 GB, Mac only) `lms get https://huggingface.co/mlx-community/Qwen2-0.5B-Instruct-4bit -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run Qwen2 0.5B Instruct (Q4_K_M, needs ~0.79 GB, from [ollama.com](https://ollama.com/library/qwen2/tags))
```sh
ollama run qwen2:0.5b-instruct-q4_K_M
```
Other versions: Q8_0 (~1.09 GB) `ollama run qwen2:0.5b-instruct-q8_0`
### MLX (Mac)
1. Install MLX (Mac). Apple silicon Macs only, text only.
```sh
# with uv
uv tool install mlx-lm
# or
pip install mlx-lm
```
2. Run Qwen2 0.5B Instruct (4bit, needs ~0.3 GB, Mac only, from [huggingface.co/mlx-community](https://huggingface.co/mlx-community/Qwen2-0.5B-Instruct-4bit/tree/107ca40780ff3f9088bac83c2f289f40f335affc))
```sh
mlx_lm.chat --model mlx-community/Qwen2-0.5B-Instruct-4bit
```
3. Or start an OpenAI-compatible server at http://localhost:8080/v1
```sh
mlx_lm.server --model mlx-community/Qwen2-0.5B-Instruct-4bit
```
Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2-0.5B-Instruct)
[interactive version](https://whatmodel.sh/model/qwen2-0-5b) · [JSON](https://whatmodel.sh/api?format=json&all=1)