Granite 4.0 1B

+ contribute

Granite 4.0 1B is an open-weights language model from IBM, released October 2025. Its published size is 1.63B parameters and a 128k-token context window. Licence: apache-2.0.

Catalogue identity notes

Comparison model from the Epoch AI internal-evaluation CSVs (Epoch release date 2025-10-28). Dense Granite 4.0 1B (Nano family, October 2025); distinct from the hybrid granite-4.0-h-1b (ollama granite4:1b-h). 1,631,750,144 parameters. ollama granite4:1b = 1b-bf16 (same digest).

40.6±0.8#523 of 719 LLMsfrom 32 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of Granite 4.0 1B, out of 828 with a score, then more from IBM.

#	model	score	where	RAM	released
613	North Micro Vision Instruct (2.4B) Cohere	41±2	local	~3.8GB	2026-08
614	CodeQwen1.5 7B (base) Alibaba	41±1	local	~6.4GB	2024-04
615	Granite 4.0 1B IBM	41±1	local	~2.4GB	2025-10
616	Mixtral 8x7B Instruct v0.1 Mistral	41±1	local	~29GB	2023-12
617	Molmo 7B-D (0924) Ai2	41±1	local	~35GB	2024-09
	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

agent tasks▁▁▃▆█▆▅▅▅▄▃▁ #232 of 300 * · 1.7% of its score · 1 test

Beats o3-mini, Ministral 3 14B Reasoning 2512, DeepSeek R1 0528 Qwen3 8B here, though they rank higher overall.

test	result	vs the others in its run	difficulty	counts	why
BFCL v3 +1	52.4% ✎	▄▄·······▄██ beats 4/6	medium · avg 61%	▇ 1.7%	blunt · at its level · several runs

coding█··········█ #2 of 4 · 25% of its score · 7 tests

Trails Granite 4.0 H 1B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
HumanEval lab	74.0% ✎	▁▁▁▁▁▃▅▅█▃▂▁ beats 137/190	very easy · avg 90%	▇▇▇ 6.6%	sharp · at its level
HumanEval+ lab	69.0% ✎	▁·▁▁▂▃▄▄█▅▄▃ beats 76/145	very easy · avg 88%	▇▇ 5.9%	sharp · at its level
BigCodeBench lab	30.2% ✎	▅····▃·█▃▅█▃ beats 5/12	medium · avg 42%	▇ 3.1%	at its level
MultiPL-E lab	32.2% ✎	▃··▄·▅▅▇▄█▃▄ beats 2/31	medium · avg 54%	▇ 3.0%	at its level
MBPP lab	65.0% ✎	▁▁▁▄▅█▄▅▃▁▂▁ beats 56/81	easy · avg 82%	▇ 2.7%	at its level
MBPP+ lab	57.0% ✎	▁··▁▁▃▄█▆▄▁▁ beats 37/120	easy · avg 73%	▇ 2.1%	at its level
CRUXEval-O lab	33.1% ✎	▂▂██▅▂▁·▁·▁▁ beats 11/51	medium · avg 51%	▇ 1.5%	blunt · at its level

chat & writing▁▁▁▃▅▄▅█▆▆▇▃ #245 of 393 · 6.0% of its score · 3 tests

Beats Mistral Small 4 (119B-A6.5B), Ministral 3 14B Reasoning 2512, Qwen2.5 Coder 32B Instruct here, though they rank higher overall. Trails Flan-T5 XXL, Granite 4.0 H 1B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
IFEval mean of four measures	79.6% ✎	█·█··█·██·██ beats 3/6	very easy · avg 87%	▇ 2.5%	typical
IFEval prompt-level strict lab	73.9% ✎	▁▁··▂··▃▇█▃▃ beats 4/26	easy · avg 81%	▇ 1.8%	at its level
IFEval instruction-level strict lab	80.8% ✎	▄▄····▄▄█▄·█ beats 2/8	very easy · avg 89%	▇ 1.7%	typical

math▁▂▃▅▆▅▄█▄▁▁▁ #316 of 398 · 28% of its score · 8 tests

Beats Athene-70B, OLMo 2 32B Instruct (0325), Gemma 2 9B IT here, though they rank higher overall. Trails Gemma 7B (base), DeepSeek R1 Distill Qwen 1.5B, Tulu 2 DPO 13B here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
GSM8K +1	73.4% ✎	█····▄▄▄···█ beats 4/6	very easy · avg 87%	▇▇▇▇▇ 12%	sharp · at its level · several runs
DeepMind Mathematics lab	34.0% ✎	▅····▅·█▃▃·▅ beats 2/10	medium · avg 54%	▇▇ 4.4%	at its level
GSM-Symbolic lab	72.3% ✎	▃▃····▃·▅▅█▃ beats 3/10	easy · avg 80%	▇ 2.9%	at its level
Minerva MATH lab	45.3% ✎	▄·▂▂·██▆·▆▄▄ beats 13/21	medium · avg 62%	▇ 2.8%	at its level
MGSM, 5 languages (en, es, fr, ja, zh) lab	37.8% ✎	█·▄▄██···█·▄ beats 4/10	medium · avg 49%	▇ 2.1%	at its level
MATH-500	44.8% ✎	▄··█▄·▄···▄▄ beats 1/6	easy · avg 80%	▇ 2.1%	at its level
AIME 2025	3.3% ✎	▄█··▄···▄··█ beats 3/6	hard · avg 33%	▇ 1.9%	sharp · noisy
OTIS mock AIME 2024–2025	0.8% ✎	▃▃▄▃▃▃▇▇█▃▂▅ beats 6/193	very hard · avg 8.5%	▇ 0.3%	sharp · too hard for it · noisy

other tests9 not in a task · 39% of its score

test	result	vs the others in its run	difficulty	counts	why
BIG-Bench Hard lab	60.4% ✎	▁▁▁▁▂▄▅▅█▃▄▁ beats 67/129	easy · avg 78%	▇▇▇ 8.3%	sharp · at its level
GPQA (subset unspecified) lab	29.9% ✎	▂▁▂▂▄▃▆▃▅█▄▃ beats 21/104	hard · avg 38%	▇▇▇ 7.0%	sharp · too hard for it
MMLU-Pro +1	33.5% ✎	█···█·██·███ beats 3/6	medium · avg 49%	▇▇ 5.8%	several runs
MMLU lab	59.4% ✎	▁▁▁▁▁▂▃▅█▃▆▃ beats 66/183	easy · avg 75%	▇▇ 5.5%	at its level
GPQA Diamond +1	24.0% ✎	▃▁▁▁▂▄▆▄▅▇█▅ beats 1/208	hard · avg 40%	▇▇ 4.7%	sharp · too hard for it · several runs
INCLUDE lab	42.1% ✎	▂····▃▂▄▄▄█▄ beats 3/36	medium · avg 54%	▇ 3.1%	typical
MMMLU lab	45.0% ✎	▂▁▁▂▃▂▄▄▅██▃ beats 8/46	medium · avg 58%	▇ 2.4%	at its level
IFBench (aggregation unspecified)	21.0% ✎	▄·█·█······█ beats 2/6	medium · avg 43%	▇ 2.0%	at its level

1 more test

test	result	vs the others in its run	difficulty	counts	why
Chess Puzzles (Epoch)	0.0% ✎	█▁▁▁▁▂▁▄▂▂▁▁ beats 11/135	very hard · avg 5.9%	▇ 0.1%	sharp · too hard for it · 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 4.0 1B · needs ~1 GB · from huggingface.co/ibm-granite
chat in the terminal
llama cli -hf ibm-granite/granite-4.0-1b-GGUF:Q4_K_M
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf ibm-granite/granite-4.0-1b-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 4.0 1B · needs ~1 GB · from huggingface.co/ibm-granite
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/ibm-granite/granite-4.0-1b-GGUF@Q4_K_M -y
Then run lms ls to see its name, and lms chat followed by that name

Ollama

Install Ollama
download it (macOS / Windows) ↗
Linux
curl -fsSL https://ollama.com/install.sh | sh
Run Granite 4.0 1B · needs ~4.41 GB · from ollama.com
chat in the terminal
ollama run granite4:1b

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 4.0 1B · needs ~0.9 GB, Mac only · from huggingface.co/mlx-community
chat in the terminal
mlx_lm.chat --model mlx-community/granite-4.0-1b-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/granite-4.0-1b-4bit

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

$ how the score was worked out · ibm-granite/granite-4.0-h-1b… 77%, LiquidAI/LFM2.5-1.2B-Thinking… 20%, Epoch AI — internal evaluatio… 3%

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-4.0-h-1b model card · lab-reported	21	▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇▇ 77%	Apache-2.0
LiquidAI/LFM2.5-1.2B-Thinking model card	8	▇▇▇▇▇▇ 20%	LFM Open License v1.0
Epoch AI — internal evaluations	3	▇ 3%	CC BY 4.0
prior · pull toward the middle		▇ 0%
weighted average		= 40.6

$ 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