Comparisons of Tiny Open Source AI Models (≤4B)

Name: Openness
Creator: Artificial Analysis
License: https://artificialanalysis.ai/docs/legal/Terms-of-Use.pdf

Open source AI models with 4B parameters or fewer. These are usually the smallest models in terms of resource demand.

Models are considered open source (also commonly referred to as open weights) where their weights are accessible to download. This allows self-hosting on your own infrastructure and enables customizing the model such as through fine-tuning.

For more details including relating to our methodology, see our FAQs.

K2 Horizon 3.7B and

MiniCPM5-2B are the highest intelligence Tiny open source models, defined as those with ≤4B parameters, followed by

G9v3-3B &

Granite 4.2 3B.

Openness

Artificial Analysis Openness Index · Higher is better

Intelligence

Artificial Analysis Intelligence Index · Higher is better

Total Parameters

Trainable parameters in billions

Openness

Artificial Analysis Openness Index: Score

Openness Index assesses model openness on a 0 to 100 normalized scale (higher is more open)

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index v4.3.2 incorporates 10 evaluations: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1

Estimate (independent evaluation forthcoming)

Artificial Analysis Intelligence Index v4.3.2 includes: AA-Briefcase v1.1, GDPval-AA v2.1, AutomationBench-AA, Terminal-Bench 4.0, SciCode, Humanity's Last Exam, GDP.pdf, CritPt, AA-Omniscience, AA-LCR v1.1. See Intelligence Index methodology for further details, including a breakdown of each evaluation and how we run them.

Intelligence Evaluations

Intelligence evaluations measured independently by Artificial Analysis · Higher is better

AA-Briefcase v1.1Updated

Agentic knowledge work, (Elo-500)/2000

GDPval-AA v2.1Updated

Agentic real-world work tasks, (Elo-500)/2000

AutomationBench-AA

Agentic SaaS workflows

Terminal-Bench 4.0

Agentic coding & terminal use

SciCodeUnder review

Coding

Humanity's Last Exam

Reasoning & knowledge

GDP.pdf

Professional document reasoning, All-pass

CritPtUnder review

Physics reasoning

AA-Omniscience Accuracy

Knowledge

AA-Omniscience Non-Hallucination Rate

1 - hallucination rate

AA-LCR v1.1

Long context reasoning

Harvey LAB-AA v1.1Updated

Legal agentic work, Hallucination-Gated All-Pass Rate

No data available

EnterpriseOps-Gym-AA

Agentic business operations

No data available

Terminal-Bench-Science 0.1New

Agentic scientific research workflows in a terminal

No data available

AA-AnalystAgent

Quantitative analysis on spreadsheets & documents

No data available

ITBench-AA

Kubernetes incident root-cause analysis

No data available

MMMU-Pro

Visual reasoning

MLCR-AA

Medical long context reasoning

No data available

While model intelligence generally translates across use cases, specific evaluations may be more relevant for certain use cases.

Size

Model Size: Total and Active Parameters

Comparison between total model parameters and parameters active during inference (billions)

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

The number of parameters actually executed during each inference forward pass, expressed in billions. For Mixture of Experts (MoE) models, a routing mechanism selects a subset of experts per token, resulting in fewer active than total parameters. Dense models use all parameters, so active equals total.

Intelligence Index vs. Active Parameters

Artificial Analysis Intelligence Index · Active parameters at inference time (billions)

Most attractive quadrant

Pareto line

Intelligence Index vs. Total Parameters

Artificial Analysis Intelligence Index · Size in parameters (billions)

Most attractive quadrant

Pareto line

The total number of trainable weights and biases in the model, expressed in billions. These parameters are learned during training and determine the model's ability to process and generate responses.

Context Window

Context window: tokens limit · Higher is better

Larger context windows are relevant to RAG (Retrieval Augmented Generation) LLM workflows which typically involve reasoning and information retrieval of large amounts of data.

Maximum number of combined input & output tokens. Output tokens commonly have a significantly lower limit (varied by model).

Further details

							Weights		Provider Benchmarks
K2 Horizon 3.7B	Institute of Foundation Models	16	3.7B	524k	-	-	🤗	-	View
MiniCPM5-2B	OpenBMB	12	2.6B	131k	-	-	🤗	-	View
G9v3-3B	AI9Stars	11	3B	131k	-	-	🤗		View
Granite 4.2 3B	IBM	9	3B	131k	$0.0	217	🤗		View
MiniCPM5-1B (Reasoning)	OpenBMB	9	1B	128k	-	-	🤗	-	View
MiniCPM5-1B (Non-reasoning)	OpenBMB	9	1B	128k	-	-	🤗	-	View
Nanbeige4.1-3B	Nanbeige	8	3.9B	256k	-	-	🤗	-	View
LFM2.5-2.6B	Liquid AI	8	2.7B	128k	-	-	🤗	-	View
Qwen3.5 2B (Reasoning)	Alibaba	7	2.3B	262k	-	-	🤗	-	View
Phi-4 Mini Instruct	Microsoft	6	3.8B	128k	-	45	🤗		View
Qwen3.5 2B (Non-reasoning)	Alibaba	6	2.3B	262k	-	-	🤗	-	View
Qwen3.5 0.8B (Reasoning)	Alibaba	6	0.9B	262k	-	-	🤗	-	View

Comparisons of Tiny Open Source AI Models (≤4B)

Openness

Intelligence

Total Parameters

Openness

Artificial Analysis Openness Index: Score

Intelligence

Artificial Analysis Intelligence Index

Artificial Analysis Intelligence Index

Intelligence Evaluations

Intelligence Evaluation Relevance

Artificial Analysis Intelligence Index

Size

Model Size: Total and Active Parameters

Total Parameters

Active Parameters at Inference Time

Intelligence Index vs. Active Parameters

Artificial Analysis Intelligence Index

Active Parameters at Inference Time

Intelligence Index vs. Total Parameters

Artificial Analysis Intelligence Index

Total Parameters

Context Window

Context Window

Context Window for RAG

Context Window

Further details