whatmodel.sh / QwQ-32B-Preview
QwQ-32B-Preview
↗ share
+ contribute
QwQ-32B-Preview is an open-weights language model from Alibaba, released November 2024. Its published size is 32.8B parameters and a 33k-token context window. Licence: apache-2.0.
Catalogue identity notes
Arena entry 'qwq-32b-preview' (text). LMArena's leaderboard table (arena-leaderboard space @6f2eeeb) labels 'qwq-32b-preview' as 'QwQ-32B-Preview' (Alibaba, Apache 2.0), linking https://huggingface.co/Qwen/QwQ-32B-Preview. Maker repo Qwen/QwQ-32B-Preview @91906fe: licence tag apache-2.0. 32,763,876,352 parameters from the Hugging Face safetensors metadata (dense, so active = total). Released 2024-11: month the maker's Hugging Face repo was created (createdAt 2024-11-27).
$ how it compares
The LLMs ranked either side of QwQ-32B-Preview, out of 828 with a score, then more from Alibaba.
| # | model | score | where | ||
|---|---|---|---|---|---|
| 526 |  OpenThinker 7B | 45±2 | local | ||
| 527 |  Phi-3 Medium 128K Instruct | 45±1 | local | ||
| 528 | QwQ-32B-Preview | 45±2 | local | ||
| 529 |  DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B) | 45±1 | local | ||
| 530 |  GPT-4o Audio Preview (2024-10-01) | 45±3 | comparison only | ||
| more from Alibaba | |||||
| 18 | Qwen3.8 Max (2.4T-A95B) | 78±1 | api $3.00/M +local | ||
| 34 | Qwen3.8 Max (0902) | 76±1 | api $3.00/M | ||
| 37 | Qwen3.8-Flash-Next | 75±1 | local | ||
$ what it's good at
agent tasks▃▁▃▆█▆▅▅▅▄▃▁ #298 of 300 * · 6.7% of its score · 1 test
Trails Granite 4.0 350M, Qwen2.5 0.5B Instruct, Granite 4.0 H 350M here, though they rank lower overall.
coding▁▁▁▃▆██▇▆▄▃▁ #155 of 458 * · 3.1% of its score · 1 test
Beats Gemini 2.5 Pro, Gemini 3.1 Flash-Lite, Qwen3-Coder-Next (80B-A3B) here, though they rank higher overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LiveCodeBench official release_v5 (Nov 1, 2023–Feb 1, 2025) | 55.8% ✎ | █···█····█·█ beats 1/3 | 3.1% |
chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #343 of 393 * · 30% of its score · 1 test
Beats Gemma 4 E2B, Code Llama 70B Instruct here, though they rank higher overall. Trails Gemma 3 270M IT, SmolLM2 360M Instruct, Granite 4.0 H 350M here, though they rank lower overall.
| test | result | vs the others in its run | counts |
|---|---|---|---|
| LMArena (text) | 1154.2 ✎ | ▁▁▁▂▃▃▃▆▆▇█▃ beats 43/379 | 30% |
other tests5 not in a task · 59% of its score
| test | result | vs the others in its run | counts |
|---|---|---|---|
| EvalPerf correctness (118 tasks, 20240328) | 83.7% ✎ | ▂·▂▂▂▆▅▆▆▆▃█ beats 22/29 | 23% |
| BRIDGE clinical text, zero-shot average (87 tasks) | 31.7% ✎ | ▁▁▁▁▁▂▂▃▄▆█▃ beats 35/97 | 14% |
| Open Persian LLM Leaderboard v2 average (21 tasks) | 61.9% ✎ | ▂▄▂▃█▄▃▆▃▅▂▂ beats 56/81 | 13% |
| Open PT-LLM Leaderboard average (9 Portuguese tasks) | 76.9% ✎ | ▃▁▁▁▁▁▁▁▂█▅▁ beats 238/276 | 6.8% |
| Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate) | 83.9% ✎ | ▁▁▃▃▄▄█▆▃▂▂▁ beats 11/140 | 2.3% |
$ where to run it
llama.cpp · recommended
version
- 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 | shWindowsirm https://llama.app/install.ps1 | iex - Run QwQ-32B-Preview · needs ~19.9 GB · from huggingface.co/lmstudio-communitychat in the terminal
llama cli -hf lmstudio-community/QwQ-32B-Preview-GGUF:Q4_K_M - Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf lmstudio-community/QwQ-32B-Preview-GGUF:Q4_K_M
LM Studio
version
- Install LM Studio · Open the app once before using the lms command.
- Run QwQ-32B-Preview · needs ~19.9 GB · from huggingface.co/lmstudio-community
- Or download it in the terminal
lms get https://huggingface.co/lmstudio-community/QwQ-32B-Preview-GGUF@Q4_K_M -y - Then run lms ls to see its name, and lms chat followed by that name
Ollama
version
- Install Ollama
download it (macOS / Windows) ↗
Linuxcurl -fsSL https://ollama.com/install.sh | sh - Run QwQ-32B-Preview · needs ~20.18 GB · from ollama.comchat in the terminal
ollama run qwq:32b-preview-q4_K_M
MLX (Mac)
version
- Install MLX (Mac) · Apple silicon Macs only, text only.with uv
uv tool install mlx-lmorpip install mlx-lm - Run QwQ-32B-Preview · needs ~18.4 GB, Mac only · from huggingface.co/mlx-communitychat in the terminal
mlx_lm.chat --model mlx-community/Qwen_QwQ-32B-Preview_MLX-4bit - Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/Qwen_QwQ-32B-Preview_MLX-4bit
Official weights and documentation: huggingface.co/Qwen ↗
$ how the score was worked out · LMArena leaderboard dataset 30%, EvalPlus / EvalPerf published… 23%, BRIDGE Medical Leaderboard (H… 14% and 5 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
| source | results | share of the score |
|---|---|---|
 LMArena leaderboard dataset | 1 | 30% |
 EvalPlus / EvalPerf published evaluation runs | 1 | 23% |
| BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab) | 1 | 14% |
| Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab) | 1 | 13% |
| Open Portuguese LLM Leaderboard (archived) | 1 | 7% |
| Berkeley Function Calling Leaderboard result archive | 1 | 7% |
| LiveCodeBench official website results (2025-08-01) | 1 | 3% |
 Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository) | 1 | 2% |
| prior · pull toward the middle | 2% | |
| weighted average | = 45.0 |
$ improve this page
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