whatmodel.sh / Qwen2.5 32B Instruct

Qwen2.5 32B Instruct

+ contribute

Qwen2.5 32B Instruct is an open-weights language model from Alibaba, released September 2024. Its published size is 32.8B parameters and a 131k-token context window. Licence: apache-2.0.

54.8±0.6#269 of 719 LLMsfrom 35 benchmark results across 17 sources

$ how it compares

The LLMs ranked either side of Qwen2.5 32B Instruct, out of 828 with a score, then more from Alibaba.

#	model	score	where	RAM	released
321	GigaChat 3.5 Ultra Instruct (432B-A28B) Sber	55±1	local	~457GB	2026-07
322	Llama 3.1 405B Instruct Meta	55±1	api $2.40/M +local	~160GB	2024-07
323	Qwen2.5 32B Instruct Alibaba	55±1	local	~21GB	2024-09
324	Command A Reasoning (08-2025) Cohere	55±2	local	~34GB
325	Bielik 11B v3.0 Instruct SpeakLeash	55±2	local	~8.1GB	2025-11
	more from Alibaba
18	Qwen3.8 Max (2.4T-A95B) Alibaba	78±1	api $3.00/M +local	~380GB	2026-08
34	Qwen3.8 Max (0902) Alibaba	76±1	api $3.00/M		2026-09
37	Qwen3.8-Flash-Next Alibaba	75±1	local	~59GB	2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #198 of 300 * · 0.8% of its score · 1 test

Beats o1 (2024-12-17), o3-mini, DeepSeek R1 (Jan '25) here, though they rank higher overall. Trails SmolLM3 3B, Jamba 1.5 Mini, Mellum2 12B-A2.5B Instruct here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
BFCL v3	62.8% ✎	▁▁▁▁▁▃▅█▅▄▁▁ beats 87/114	medium · avg 52%	▇ 0.8%	blunt · at its level

coding▁▁▁▃▆██▇▆▄▃▁ #196 of 458 · 20% of its score · 6 tests

Beats Gemini 2.5 Flash, Qwen3.5 4B, Qwen3 30B-A3B Instruct 2507 here, though they rank higher overall. Trails QwQ-32B-Preview, Ministral 3 3B Reasoning 2512, DeepSeek-Coder-V2 Instruct (0724) here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
LiveCodeBench official (Jan 1–Nov 1, 2024)	40.1% ✎	▁▁·▁▂▄▆█▅▂·▂ beats 24/31	hard · avg 26%	▇▇▇▇ 9.1%	sharp · at its level
CRUXEval (ZeroEval, 800 problems)	69.9% ✎	▁▂▆▃▆█▆▃▂·▁▁ beats 28/47	medium · avg 61%	▇▇▇ 7.9%	sharp · at its level
CodeReviewQA: change localisation (hard), invariant accuracy	67.6% ✎	█▂▁▁▂▂▂▃▅▅▅▇ beats 46/47	medium · avg 36%	▇ 1.2%	sharp · at its level · noisy
CodeReviewQA: solution identification (hard), invariant accuracy	89.5% ✎	▃··▁▂▂▃▂█▃▄▃ beats 46/47	medium · avg 54%	▇ 0.8%	sharp · at its level · noisy
CodeReviewQA: automated code refinement, exact match rate	45.0% ✎	▁▁▁▁▁▁▂▂▃█▅▄ beats 42/47	hard · avg 33%	▇ 0.8%	at its level
CodeReviewQA: change type recognition, invariant accuracy	71.0% ✎	▁▁▁······▁▃█ beats 20/47	easy · avg 67%	▇ 0.0%	blunt · at its level · noisy

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #225 of 393 · 5.8% of its score · 3 tests

Beats Granite 4.2 8B, Mistral Small 4 (119B-A6.5B), Olmo 3.1 32B Think here, though they rank higher overall. Trails Flan-T5 XXL, LFM2.5 1.2B Instruct, Gemma 3 4B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
Arabic IFEval prompt-level (Inception)	60.4% ✎	▁·▁▁▂▁▄▄▆▄█▄ beats 34/38	medium · avg 49%	▇ 2.8%	sharp · at its level
AlpacaEval 2 length-controlled	39.1% ✎	▃▃·▁▁█▇▅·▁·▁ beats 14/22	medium · avg 42%	▇ 2.4%	at its level
IFEval prompt-level loose	82.4% ✎	▁▃▁▃▁▄▁▁█▃▃▁ beats 12/22	easy · avg 80%	▇ 0.6%	typical

math▁▂▃▅▆▅▄█▄▁▁▁ #231 of 398 · 20% of its score · 6 tests

Beats GPT-4o (2024-08-06), Command A (03-2025), GPT-4o (2024-11-20) here, though they rank higher overall. Trails OpenThinker 7B, Olmo 3 7B Instruct, Tulu 2 70B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
MATH level 5	56.1% ✎	▂▂▅▆▄█▆▃▅▂▃▄ beats 48/92	hard · avg 32%	▇▇▇ 8.7%	sharp · at its level
MATH Level 5 (ZeroEval, 721 problems)	59.5% ✎	▁▁▁▆█▂▇▇▄··▂ beats 33/38	hard · avg 34%	▇▇ 5.3%	sharp · at its level
OTIS mock AIME 2024–2025	7.4% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 42/193	very hard · avg 8.5%	▇ 2.3%	sharp · noisy
MATH (subset unspecified) +1	77.9% ✎	▃▃█▂▂▆·····▁ beats 21/22	medium · avg 61%	▇ 1.7%	at its level · several runs
GSM8K	87.5% ✎	▂▃··▃▂▅█▆▃▂▃ beats 15/22	very easy · avg 88%	▇ 1.1%	sharp · maxed out
OMEGA (math generalisation)	19.2% ✎	▅···▃▃···▃▃█ beats 2/8	hard · avg 24%	▇ 1.1%	at its level

other tests18 not in a task · 53% of its score

test	result	vs the others in its run	difficulty	counts	why
MMLU-Redux (ZeroEval, 2,778 questions)	83.7% ✎	▁·▁▁▄▄▄▅█▄▂▁ beats 41/53	easy · avg 77%	▇▇▇▇▇ 12%	typical
ZebraLogic (ZeroEval, 1,000 puzzles)	26.1% ✎	▁▁▃▆██▁▁·▁▁▁ beats 40/57	hard · avg 18%	▇▇▇ 7.5%	sharp · at its level
GPQA Diamond	46.1% ✎	▁▁▁▁▂▄▆▄▅▇█▅ beats 53/208	hard · avg 40%	▇▇ 5.6%	sharp · at its level
Medmarks-V mean win rate (19 verifiable medical benchmarks, 71 configurations)	0.5003 ✎	▁▄▃▃▇▄█▃▅▁▁▁ beats 16/30	avg 0.423	▇▇ 5.3%	sharp
BIG-Bench Hard	82.3% ✎	▁▄·▁▁·▅█·▄▄▁ beats 18/22	easy · avg 75%	▇ 3.6%	sharp
MMLU	84.7% ✎	▁▂·▁·▁▁█·▃▃▅ beats 18/22	easy · avg 77%	▇ 3.5%	typical
Medmarks-V weighted mean win rate (61 models, 71 configurations)	50.0% ✎	▂▃▅▂▆▃█▅█▅▃▆ beats 16/34	medium · avg 42%	▇ 2.6%	blunt · at its level
CPTUB average (SpeakLeash, 0-5 judge scale)	75.0% ✎	▁▂▂▂▃█▃▆▅▄▂▃ beats 70/95	easy · avg 66%	▇ 2.1%	at its level

10 more tests

test	result	vs the others in its run	difficulty	counts	why
BRIDGE clinical text, zero-shot average (87 tasks)	39.9% ✎	▁▁▁▁▁▂▂▃▄▆█▃ beats 73/97	hard · avg 32%	▇ 1.8%	blunt · at its level
EvalPerf correctness (118 tasks, 20240328)	87.6% ✎	▂·▂▂▂▆▅▆▆▆▃█ beats 25/29	easy · avg 76%	▇ 1.7%	typical
Polish EQ-Bench v2 (SpeakLeash)	71.2% ✎	▁··▁▁▂▂▆▅▆█▆ beats 56/70	medium · avg 64%	▇ 1.6%	at its level
Open Persian LLM Leaderboard v2 average (21 tasks)	62.8% ✎	▂▄▂▃█▄▃▆▃▅▂▂ beats 57/81	medium · avg 58%	▇ 1.4%	blunt · at its level
PopQA	26.1% ✎	▁▁▁▃▃▃▃█▁▃·▄ beats 9/21	hard · avg 32%	▇ 1.0%	blunt · at its level
SurgiQ surgical MCQ (zero-shot, log-probability scoring)	64.9% ✎	▂···▂▂·▃▂▃██ beats 15/17	medium · avg 53%	▇ 0.8%	at its level
TruthfulQA MC2	70.6% ✎	▃▃▂·▃▃·▅█▂▅▂ beats 19/21	medium · avg 63%	▇ 0.7%	blunt · at its level
Open PT-LLM Leaderboard average (9 Portuguese tasks)	80.2% ✎	▃▁▁▁▁▁▁▁▂█▅▁ beats 261/276	medium · avg 73%	▇ 0.7%	at its level · noisy
Chess Puzzles (Epoch)	0.0% ✎	█▁▁▁▁▂▁▄▂▂▁▁ beats 40/135	very hard · avg 5.9%	▇ 0.3%	sharp · too hard for it · noisy
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)	97.0% ✎	▁▁▂▃▄▄█▆▃▂▂▁ beats 104/140	very easy · avg 94%	▇ 0.2%	blunt · maxed out · noisy

$ 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 | sh
Windows
irm https://llama.app/install.ps1 | iex
other ways to install ↗
Run Qwen2.5 32B Instruct · needs ~19.9 GB · from huggingface.co/Qwen (official)
chat in the terminal
llama cli -hf Qwen/Qwen2.5-32B-Instruct-GGUF:q4_k_m
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf Qwen/Qwen2.5-32B-Instruct-GGUF:q4_k_m

LM Studio

version

Install LM Studio · Open the app once before using the lms command.
download it (macOS / Windows / Linux) ↗
Run Qwen2.5 32B Instruct · needs ~19.9 GB · from huggingface.co/Qwen (official)
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/Qwen/Qwen2.5-32B-Instruct-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) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Qwen2.5 32B Instruct · needs ~20.18 GB · from ollama.com
chat in the terminal
ollama run qwen2.5:32b

MLX (Mac)

version

Install MLX (Mac) · Apple silicon Macs only, text only.
with uv
uv tool install mlx-lm
or
pip install mlx-lm
Run Qwen2.5 32B Instruct · needs ~18.4 GB, Mac only · from huggingface.co/lmstudio-community
chat in the terminal
mlx_lm.chat --model lmstudio-community/Qwen2.5-32B-Instruct-MLX-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model lmstudio-community/Qwen2.5-32B-Instruct-MLX-4bit

Official weights and documentation: huggingface.co/Qwen ↗

$ how the score was worked out · WildEval ZeroEval published r… 33%, Epoch AI — internal evaluatio… 17%, 2 OLMo 2 Furious (Team OLMo,… 14% and 14 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	licence
WildEval ZeroEval published results	4	▇▇▇▇▇▇▇▇▇▇ 33%	Apache-2.0
Epoch AI — internal evaluations	4	▇▇▇▇▇ 17%	CC BY 4.0
2 OLMo 2 Furious (Team OLMo, Ai2) — authors' own evaluations of OLMo 2 Instruct and peer instruct models	8	▇▇▇▇ 14%	CC BY 4.0
LiveCodeBench official website results (2024-12-26)	1	▇▇▇ 9%	CC-BY-SA-4.0
Medmarks: A Comprehensive Open-Source LLM Benchmark Suite for Medical Tasks — authors' own runs of 61 models	2	▇▇ 8%	CC BY 4.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)	4	▇ 3%	CC-BY-4.0
Inception Arabic/Hindi leaderboards (AraGen-v3, HindiGen-v1, Arabic IFEval)	1	▇ 3%	Apache-2.0
SpeakLeash CPTUB (Complex Polish Text Understanding Benchmark)	1	▇ 2%	MIT
BRIDGE Medical Leaderboard (Harvard Medical School / Brigham and Women's YLab)	1	▇ 2%	MIT
EvalPlus / EvalPerf published evaluation runs	1	▇ 2%	Apache-2.0
allenai/Olmo-3.1-32B-Think model card	2	▇ 2%	Apache-2.0
SpeakLeash Polish EQ-Bench leaderboard	1	▇ 2%	Apache-2.0
Open Persian LLM Leaderboard v2 (Part DP AI and AUT NLP Lab)	1	▇ 1%	Apache-2.0
SurgiQ: A Large-Scale Multi-Domain Benchmark for Evaluating Surgical Understanding in Large Language Models (arXiv:2606.08071v1)	1	▇ 1%	CC BY 4.0
Berkeley Function Calling Leaderboard result archive	1	▇ 1%	Apache-2.0
Open Portuguese LLM Leaderboard (archived)	1	▇ 1%	Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇ 0%	Apache-2.0
prior · pull toward the middle		▇ 0%
weighted average		= 54.8