whatmodel.sh / Falcon3 3B Instruct

Falcon3 3B Instruct

+ contribute

Falcon3 3B Instruct is an open-weights language model from TII, released December 2024. Its published size is 3.23B parameters and a 32k-token context window. Licence: falcon-llm.

Catalogue identity notes

Maker repo tiiuae/Falcon3-3B-Instruct @411bb94: licence tag other (falcon-llm-license). 3,227,655,168 parameters from the Hugging Face safetensors metadata. Dense, so active = total. Released 2024-12: month the maker's Hugging Face repo was created (createdAt 2024-12-14). Ollama tags falcon3:3b-instruct-q4_K_M, falcon3:3b-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.

36.5±3.1#584 of 719 LLMsfrom 7 benchmark results across 3 sources

$ how it compares

The LLMs ranked either side of Falcon3 3B Instruct, out of 828 with a score, then more from TII.

#	model	score	where	RAM	released
675	Qwen1.5 7B Chat Alibaba	37±1	local	~6.5GB	2024-02
676	Yi 1.5 6B Chat 01.AI	37±2	local	~5.1GB	2024-05
677	Falcon3 3B Instruct TII	37±3	local	~3.4GB	2024-12
678	Zephyr 7B alpha Hugging Face	36±1	local	~6.1GB	2023-10
679	Vicuna 13B v1.5 LMSYS	36±2	local	~9.8GB	2023-08
	more from TII
233	Falcon-H1R 7B TII	60±1	local	~6GB	2026-01
332	Falcon-H1 34B Instruct TII	54±3	local	~23GB	2025-05
494	Falcon-H1 7B Instruct TII	46±3	local	~6GB	2025-05

see all LLMs · find the ones that fit your machine

$ what it's good at

agent tasks▁▃▃▆█▆▅▅▅▄▃▁ #293 of 300 * · 43% of its score · 2 tests

Beats QwQ-32B-Preview, DeepSeek-Coder-V2 Lite Instruct (16B-A2.4B), Phi-4-mini 3.8B here, though they rank higher overall. Trails Granite 4.0 350M, Qwen2.5 0.5B Instruct, Granite 4.0 H 350M here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
BFCL V4 (December 2025 archive)	16.3% ✎	▁▁▃▂██▅▅▆▂▂▂ beats 3/77	hard · avg 32%	▇▇▇▇▇▇▇▇▇▇ 24%	typical
BFCL v3	32.5% ✎	▁▁▁▃▁▃▅█▅▄▁▁ beats 8/114	medium · avg 52%	▇▇▇▇▇▇▇ 18%	blunt · at its level

coding▁▁▁▃▆██▇▆▄▃▁ #375 of 458 · 36% of its score · 4 tests

Beats LFM2.5 8B-A1B, DeepSeek R1 Distill Qwen 7B, Qwen3.5 2B here, though they rank higher overall. Trails Qwen2.5 1.5B Instruct, WizardLM 13B V1.2 here, though they rank lower overall.

test	result	vs the others in its run	difficulty	counts	why
CodeReviewQA: solution identification (hard), invariant accuracy	19.5% ✎	▃··▁▂▂▃▂█▃▄▂ beats 14/47	medium · avg 54%	▇▇▇▇▇▇ 14%	sharp · noisy
CodeReviewQA: automated code refinement, exact match rate	14.2% ✎	▁▁▁▁▁▁▃▂▃█▅▄ beats 8/47	hard · avg 33%	▇▇▇▇▇ 12%	typical
CodeReviewQA: change localisation (hard), invariant accuracy	1.7% ✎	█▂▃▁▂▂▂▃▅▅▅▇ beats 11/47	medium · avg 36%	▇▇▇▇ 10%	sharp · noisy
CodeReviewQA: change type recognition, invariant accuracy	52.0% ✎	▁▁▁······▃▃█ beats 5/47	easy · avg 67%	▇ 0.4%	blunt · at its level · noisy

other tests1 not in a task · 17% of its score

test	result	vs the others in its run	difficulty	counts	why
Open PT-LLM Leaderboard average (9 Portuguese tasks)	64.1% ✎	▃▁▁▁▁▁▁▁▂█▅▁ beats 107/276	medium · avg 73%	▇▇▇▇▇▇▇ 17%	at its level · 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 Falcon3 3B Instruct · needs ~2 GB · from huggingface.co/tiiuae (official)
chat in the terminal
llama cli -hf tiiuae/Falcon3-3B-Instruct-GGUF:q4_k_m
Or start a local server and open the address it prints (http://localhost:9931)
llama serve -hf tiiuae/Falcon3-3B-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 Falcon3 3B Instruct · needs ~2 GB · from huggingface.co/tiiuae (official)
open in LM Studio ↗
Or download it in the terminal
lms get https://huggingface.co/tiiuae/Falcon3-3B-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 Falcon3 3B Instruct · needs ~2.44 GB · from ollama.com
chat in the terminal
ollama run falcon3:3b-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 Falcon3 3B Instruct · needs ~1.8 GB, Mac only · from huggingface.co/mlx-community
chat in the terminal
mlx_lm.chat --model mlx-community/Falcon3-3B-Instruct-4bit
Or start an OpenAI-compatible server at http://localhost:8080/v1
mlx_lm.server --model mlx-community/Falcon3-3B-Instruct-4bit

Official weights and documentation: huggingface.co/tiiuae ↗

$ how the score was worked out · Berkeley Function Calling Lea… 43%, CodeReviewQA: The Code Review… 36%, Open Portuguese LLM Leaderboa… 17%

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
Berkeley Function Calling Leaderboard result archive	2	▇▇▇▇▇▇▇▇▇▇▇▇▇ 43%	Apache-2.0
CodeReviewQA: The Code Review Comprehension Assessment for Large Language Models (arXiv:2503.16167v2)	4	▇▇▇▇▇▇▇▇▇▇▇ 36%	CC-BY-4.0
Open Portuguese LLM Leaderboard (archived)	1	▇▇▇▇▇ 17%	Apache-2.0
prior · pull toward the middle		▇ 4%
weighted average		= 36.5

$ 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