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

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.

Institute of Foundation Models logoK2 Horizon 3.7B and OpenBMB logoMiniCPM5-2B score highest on the Artificial Analysis Progress Index v1 among Tiny open source models, defined as those with ≤4B parameters, followed by Alibaba logoQwen3 4B 2507 & IBM logoGranite 4.2 3B.
Artificial Analysis Openness Index · Higher is better
Artificial Analysis Progress Index v1 · Higher is better
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 Progress Index

Artificial Analysis Progress Index v1 is an item response theory estimate of general capability across Artificial Analysis evaluations

Evaluations

Evaluations measured independently by Artificial Analysis · AA-PersonalityBench scores are percentile ranks from 1 to 100
See more

Agentic knowledge work, Elo

Agentic SaaS workflows

Agentic scientific research workflows in a terminal

Agentic software engineering in open-source repositories

Reasoning & knowledge

Professional document reasoning, All-pass

Agentic coding & terminal use

Physics reasoning

Quantitative analysis on spreadsheets & documents

No data available

Legal agentic work, Hallucination-Gated All-Pass Rate

No data available

Agentic business operations

No data available

Agentic real-world work tasks, Elo

Agentic interactive 3D experiences, Elo

No data available

Product and brand videos designed in code, Elo

No data available

Kubernetes incident root-cause analysis

No data available

Medical long context reasoning

No data available

Visual reasoning

Rich, varied, well-crafted language

No data available

Well-structured, scannable, answer-first

No data available

Truthful, acknowledges uncertainty, non-sycophantic

No data available

Use of filler and generic language

No data available

Size

Model Size: Total and Active Parameters

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

Progress Index vs. Active Parameters

Artificial Analysis Progress Index v1 · Active parameters at inference time (billions)
Most attractive quadrant
Pareto line

Progress Index vs. Total Parameters

Artificial Analysis Progress Index v1 · Size in parameters (billions)
Most attractive quadrant
Pareto line

Context Window

Context Window

Context window: tokens limit · Higher is better

Further details

Weights
Provider Benchmarks
K2 Horizon 3.7B
Institute of Foundation Models logoInstitute of Foundation Models
242
3.7B
524k
-
-
-
-
MiniCPM5-2B
OpenBMB logoOpenBMB
221
2.6B
131k
-
-
-
-
Granite 4.2 3B
IBM logoIBM
191
3B
131k
-
-
-
-
G9v3-3B
AI9Stars logoAI9Stars
189
3B
131k
-
-
-
-
Nanbeige4.1-3B
Nanbeige logoNanbeige
180
3.9B
256k
-
-
-
-
Qwen3.5 2B (Reasoning)
Alibaba logoAlibaba
177
2.3B
262k
-
-
-
-
NVIDIA Nemotron 3 Nano 4B
NVIDIA logoNVIDIA
176
4B
262k
-
-
-
-
LFM2.5-2.6B
Liquid AI logoLiquid AI
171
2.7B
128k
-
-
-
-
Exaone 4.0 1.2B (Reasoning)
LG AI Research logoLG AI Research
171
1.3B
64k
-
-
-
-
Qwen3.5 2B (Non-reasoning)
Alibaba logoAlibaba
171
2.3B
262k
-
-
-
-
MiniCPM5-1B (Reasoning)
OpenBMB logoOpenBMB
167
1B
128k
-
-
-
-
MiniCPM5-1B (Non-reasoning)
OpenBMB logoOpenBMB
163
1B
128k
-
-
-
-