[whatmodel.sh](../../README.md) / Qwen2.5 0.5B Instruct # Qwen2.5 0.5B Instruct Alibaba · released 2024-09 · 0.49B params · apache-2.0 **score 29.8** ±1.7 · #662 of 732 · from 12 benchmark results across 7 sources Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only. ## What it's good at ● marks Qwen2.5 0.5B Instruct among every model on that task or test, worst to best. 
task            vs the others          rank  counts  tests
agent tasks     ▁▁●▆█▄▅▅▄▄▂▁   #303 of 313*    5.2%      1
chat & writing  ▁▁▁▃●▄▆█▆▇▇▃   #361 of 402*    9.3%      2
math            █··········●       #1 of 2*     18%      1
agent tasks · #303 of 313 · 5.2% of its score Beats [QwQ-32B-Preview](qwq-32b-preview.md), [DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B)](deepseek-coder-v2-lite-instruct.md), [Phi-4-mini 3.8B](phi-4-mini.md) here, though they rank higher overall. Trails [Granite 4.0 350M](granite-4-0-350m.md), [Granite 4.0 H 350M](granite-4-0-h-350m.md) here, though they rank lower overall. 
test     result  vs the others in its run   difficulty        counts  why
BFCL v3   35.6%  ▁▁▁▁●▃▅█▅▄▁▁ beats 10/114  medium · avg 51%    5.2%  blunt · at its level
chat & writing · #361 of 402 · 9.3% of its score Beats [Code Llama 70B Instruct](codellama-70b-instruct.md), [Gemma 4 E2B](gemma-4-e2b.md), [Phi-3-mini-4k-instruct (June 2024 update)](phi-3-mini-4k-instruct-june-2024.md) here, though they rank higher overall. Trails [Gemma 3 270M IT](gemma-3-270m.md), [SmolLM2 360M Instruct](smollm2-360m.md), [Granite 4.0 H 350M](granite-4-0-h-350m.md) here, though they rank lower overall. 
test                                result  vs the others in its run  difficulty        counts  why
Arabic IFEval prompt-level (Incep…   13.9%  ▁·▁▁▂▁●▄▆▄█▄ beats 7/38   medium · avg 46%    5.4%  sharp · too hard for it
IFEval (aggregation unspecified)     31.6%  ●▂▄▂▂▄▆██▅▂▁ beats 0/145  easy · avg 76%      3.9%  at its level · small run
math · #1 of 2 · 18% of its score 
test   result  vs the others in its run   difficulty      counts  why
GSM8K   26.8%  ▂▂▂▂▃●▅▇██▆▄ beats 56/206  easy · avg 82%     18%  sharp · small run
other tests (8) 
test                                result  vs the others in its run   difficulty           counts  why
CPTUB average (SpeakLeash, 0-5 ju…   28.0%  ●▂▂▂▃█▃▆▅▄▂▃ beats 0/95    medium · avg 65%        20%  at its level
BIG-Bench Hard                       30.7%  ▁▁▁▁●▄▅▆█▃▃▁ beats 12/155  medium · avg 62%        16%  sharp · too hard for it · small run
Open Persian LLM Leaderboard v2 a…   33.3%  ▂●▂▃█▄▃▆▃▅▂▂ beats 4/81    medium · avg 57%        12%  blunt · at its level
HellaSwag                            48.0%  ▁▁▁▂▂●▄▆█▇▄▂ beats 21/135  easy · avg 83%         5.0%  at its level · small run
Open PT-LLM Leaderboard average (…   50.9%  ▃▁▁▁▁▁▁●▂█▅▁ beats 71/276  medium · avg 72%       4.4%  noisy
Vectara hallucination leaderboard…   74.8%  ●▁▂▃▄▄█▆▃▂▂▁ beats 2/140   very easy · avg 94%    4.2%  blunt · noisy
AraGen-v3 3C3H score (Inception, …    5.2%  ▁▁·▁●▂▄█▄▅▄▄ beats 4/66    hard · avg 29%         3.1%  too hard for it
PIQA                                 67.2%  ▁▁▁●▂▃▅█▅▃▁▁ beats 13/123  easy · avg 85%         2.0%  blunt · at its level · small run
## How the score was worked out Each result implies a score on its own, after adjusting for how hard that test is and who ran it. The model's 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 is added up by who ran the tests. [Method](../method.md) 
source                                    results  share of the score
HuggingFaceTB/SmolLM2-360M-Instruct mod…        5   45% ▇▇▇▇▇
SpeakLeash CPTUB                                1   20% ▇▇
Open Persian LLM Leaderboard v2                 1   12% ▇
Inception Arabic/Hindi leaderboards             2    9% ▇
Berkeley Function Calling Leaderboard r…        1    5% ▇
Open Portuguese LLM Leaderboard                 1    4% ▇
Vectara hallucination leaderboard               1    4% ▇
prior: pull toward the middle                        1% ▇
weighted average                                   = 29.8
Licences: HuggingFaceTB/SmolLM2-360M-Instruct model card ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) · SpeakLeash CPTUB ([MIT](../../licenses/hf-leaderboard-spaces-MIT.txt)) · Open Persian LLM Leaderboard v2 ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Inception Arabic/Hindi leaderboards ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Berkeley Function Calling Leaderboard result archive ([Apache-2.0](../../licenses/bfcl-result-archive-v4-Apache-2.0.txt)) · Open Portuguese LLM Leaderboard ([Apache-2.0](../../licenses/hf-leaderboard-spaces-Apache-2.0.txt)) · Vectara hallucination leaderboard ([Apache-2.0](https://www.apache.org/licenses/LICENSE-2.0)) ## 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 Qwen2.5 0.5B Instruct (Q4_K_M, needs ~0.5 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF/tree/9217f5db79a29953eb74d5343926648285ec7e67)) ```sh 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) ```sh llama serve -hf Qwen/Qwen2.5-0.5B-Instruct-GGUF:q4_k_m ``` Other versions: Q8_0 (~0.7 GB) `llama cli -hf Qwen/Qwen2.5-0.5B-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 Qwen2.5 0.5B Instruct (Q4_K_M, needs ~0.5 GB, from [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF/tree/9217f5db79a29953eb74d5343926648285ec7e67)) ```sh lms get https://huggingface.co/Qwen/Qwen2.5-0.5B-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 (~0.7 GB) `lms get https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct-GGUF@q8_0 -y` · 4bit (~0.3 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-4bit -y` · 8bit (~0.5 GB, Mac only) `lms get https://huggingface.co/lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-8bit -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 Qwen2.5 0.5B Instruct (Q4_K_M, needs ~0.79 GB, from [ollama.com](https://ollama.com/library/qwen2.5/tags)) ```sh ollama run qwen2.5:0.5b-instruct-q4_K_M ``` Other versions: Q8_0 (~1.09 GB) `ollama run qwen2.5:0.5b-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 Qwen2.5 0.5B Instruct (4bit, needs ~0.3 GB, Mac only, from [huggingface.co/lmstudio-community](https://huggingface.co/lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-4bit/tree/5d8965c688e53205dee290e3e3be0f2362d4535c)) ```sh 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 ```sh mlx_lm.server --model lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-4bit ``` Other versions: 8bit (~0.5 GB, Mac only) `mlx_lm.chat --model lmstudio-community/Qwen2.5-0.5B-Instruct-MLX-8bit` Model repository and documentation: [huggingface.co/Qwen](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) [interactive version](https://whatmodel.sh/model/qwen2-5-0-5b) · [JSON](https://whatmodel.sh/api?format=json&all=1)