whatmodel.sh / Granite 3.2 2B Instruct

Granite 3.2 2B Instruct

+ contribute

Granite 3.2 2B Instruct is an open-weights language model from IBM, released February 2025. Its published size is 2.53B parameters and a 131k-token context window. Licence: apache-2.0.

Catalogue identity notes

Maker repo ibm-granite/granite-3.2-2b-instruct @641593c: licence tag apache-2.0. 2,533,531,648 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2025-02: month the maker's Hugging Face repo was created (createdAt 2025-02-17). Ollama tags granite3.2:2b-instruct-q4_K_M, granite3.2:2b-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.

36.0±1.3#591 of 719 LLMsfrom 10 benchmark results across 2 sources

$ how it compares

The LLMs ranked either side of Granite 3.2 2B Instruct, out of 828 with a score, then more from IBM.

#	model	score	where	RAM	released
683	Code Llama 7B Instruct Meta	36±1	local	~5.5GB	2023-08
684	LFM2 700M Liquid AI	36±2	local	~1.9GB	2025-07
685	Granite 3.2 2B Instruct IBM	36±1	local	~3GB	2025-02
686	Mistral 7B Instruct v0.2 Mistral	36±1	local	~6GB	2023-12
687	WizardLM 13B V1.2 Microsoft	36±1	local	~9.8GB	2023-07
	more from IBM
192	Granite 4.2 30B IBM	63±1	api $0.28/M +local	~19GB	2026-08
239	Granite 4.2 8B IBM	60±1	api $0.11/M +local	~6.6GB	2026-08
276	Granite 4.2 3B IBM	57±1	api $0.052/M +local	~3.7GB	2026-08

see all LLMs · find the ones that fit your machine

$ what it's good at

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #333 of 393 * · 12% of its score · 2 tests

Beats OpenThinker3 7B, Gemma 4 E2B, Code Llama 70B Instruct here, though they rank higher overall. Trails Granite 4.0 H 350M, LFM2 350M, Gemma 3 1B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
IFEval (aggregation unspecified) lab	61.6% ✎	▄▄·▄······▄█ beats 0/5	easy · avg 79%	▇▇▇ 6.5%	at its level
Arena-Hard (version/judge unspecified) lab	24.9% ✎	█·▄··▄·····█ beats 1/5	medium · avg 58%	▇▇ 5.9%	at its level

math▁▂▃▅▆▅▄█▄▁▁▁ #354 of 398 · 41% of its score · 3 tests

Beats Yi 34B Chat, Llama 3 8B Instruct, Command R (03-2024) here, though they rank higher overall. Trails Gemma 3 1B, Qwen3 0.6B, Tulu 2 DPO 7B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
GSM8K lab	67.0% ✎	█··▄·····▄·█ beats 0/5	very easy · avg 90%	▇▇▇▇▇▇▇▇▇▇ 24%	sharp · at its level
MATH-500 lab	35.5% ✎	█···▄·▄▄···▄ beats 1/5	easy · avg 74%	▇▇▇▇ 9.5%	at its level
AIME 2024 lab	0.9% ✎	█···▄▄▄····▄ beats 0/5	very hard · avg 11%	▇▇▇ 7.3%	sharp · too hard for it

other tests5 not in a task · 46% of its score

test	result	vs the others in its run	difficulty	counts	why
BIG-Bench Hard lab	61.4% ✎	█·▄·····▄▄·▄ beats 0/5	easy · avg 84%	▇▇▇▇▇▇▇ 17%	sharp · at its level
MMLU lab	57.2% ✎	▄█········▄█ beats 2/5	easy · avg 78%	▇▇▇▇▇ 13%	at its level
DROP lab	23.8% ✎	▃▃·····▃···█ beats 1/5	easy · avg 68%	▇▇▇▇ 10%	at its level
PopQA lab	20.6% ✎	▄·█······▄·█ beats 2/5	hard · avg 34%	▇▇ 4.4%	blunt
Vectara hallucination leaderboard, 2025 dataset (HHEM-2.3, factual consistency rate)	83.5% ✎	▁▁▃▃▄▄█▆▃▂▂▁ beats 9/140	very easy · avg 94%	▇ 1.6%	blunt · 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 Granite 3.2 2B Instruct · needs ~1.5 GB · from huggingface.co/lmstudio-community
chat in the terminal
llama cli -hf lmstudio-community/granite-3.2-2b-instruct-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf lmstudio-community/granite-3.2-2b-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 Granite 3.2 2B Instruct · needs ~1.5 GB · from huggingface.co/lmstudio-community
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/lmstudio-community/granite-3.2-2b-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 Granite 3.2 2B Instruct · needs ~2.02 GB · from ollama.com
chat in the terminal
ollama run granite3.2:2b-instruct-q4_K_M

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 Granite 3.2 2B Instruct · needs ~1.4 GB, Mac only · from huggingface.co/alexgusevski
chat in the terminal
mlx_lm.chat --model alexgusevski/granite-3.2-2b-instruct-q4-mlx
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model alexgusevski/granite-3.2-2b-instruct-q4-mlx

Weights and documentation: huggingface.co/ibm-granite ↗

$ how the score was worked out · ibm-granite/granite-3.3-8b-in… 98%, Vectara hallucination leaderb… 2%

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
ibm-granite/granite-3.3-8b-instruct model card · lab-reported	9	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 98%	Apache-2.0
Vectara hallucination leaderboard (vectara/hallucination-leaderboard repository)	1	▇ 2%	Apache-2.0
prior · pull toward the middle		▇ 1%
weighted average		= 36.0

$ improve this page

Update model information Add a place to run it Add a test result Propose a benchmark Correct a benchmark score Edit 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 2026-10-11 02:20 UTC · sources and licences · json