AI Battle > Rankings > Artificial Analysis Openness Index: Score

Artificial Analysis Openness Index: Score Ranking — eval openness (2026)

K2 Think V2 by MBZUAI Institute of Foundation Models leads the Artificial Analysis Openness Index: Score ranking with a score of 88.9. The ranking covers 20 evaluated eval openness from a public dataset that AI Battle re-checks daily.

Last updated: . Source: public benchmark datasets and provider pricing, aggregated daily; 20 entries shown.

K2 Think V288.9K2-V2 (high)88.9K2-V2 (medium)88.9K2-V2 (low)88.9Olmo 3.1 32B Think88.9Olmo 3.1 32B Instruct88.9Olmo 3 7B Think88.9Molmo 7B-D88.9Olmo 3 7B Instruct88.9Apertus 70B Instruct88.9Apertus 8B Instruct88.9Nemotron 3 Ultra 550B A55B (Reasoning)83.3Nemotron 3.5 Lightning83.3Nemotron 3 Super 120B A12B (Reasoning)83.3Nemotron Cascade 2 30B A3B83.3NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)83.3MiniCPM5-1B (Reasoning)83.3MiniCPM5-1B (Non-reasoning)83.3NVIDIA Nemotron 3 Nano 4B83.3NVIDIA Nemotron 3 Nano 30B A3B (Non-reasonin83.3
Every entry in the dataset, scaled from the smallest to the largest published value.

All 20 entries in this dataset

#EntryCreatorValueΔ vs #1
1K2 Think V2MBZUAI Institute of Foundation Models88.90.0%
2K2-V2 (high)MBZUAI Institute of Foundation Models88.90.0%
3K2-V2 (medium)MBZUAI Institute of Foundation Models88.90.0%
4K2-V2 (low)MBZUAI Institute of Foundation Models88.90.0%
5Olmo 3.1 32B ThinkAllen Institute for AI88.90.0%
6Olmo 3.1 32B InstructAllen Institute for AI88.90.0%
7Olmo 3 7B ThinkAllen Institute for AI88.90.0%
8Molmo 7B-DAllen Institute for AI88.90.0%
9Olmo 3 7B InstructAllen Institute for AI88.90.0%
10Apertus 70B InstructSwiss AI Initiative88.90.0%
11Apertus 8B InstructSwiss AI Initiative88.90.0%
12Nemotron 3 Ultra 550B A55B (Reasoning)NVIDIA83.3-6.3%
13Nemotron 3.5 LightningNVIDIA83.3-6.3%
14Nemotron 3 Super 120B A12B (Reasoning)NVIDIA83.3-6.3%
15Nemotron Cascade 2 30B A3BNVIDIA83.3-6.3%
16NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)NVIDIA83.3-6.3%
17MiniCPM5-1B (Reasoning)OpenBMB83.3-6.3%
18MiniCPM5-1B (Non-reasoning)OpenBMB83.3-6.3%
19NVIDIA Nemotron 3 Nano 4BNVIDIA83.3-6.3%
20NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)NVIDIA83.3-6.3%

How the values are distributed

Across the 20 numeric entries in this window the median is 88.9 and the mean is 86.4, so the typical entry sits 0.00 below the leader. The window spans 83.3 to 88.9.

900000011
8 equal buckets between the smallest (83.3) and largest (88.9) value; numbers under bars are entry counts.

11 of 20 entries are at or above the median, and the top three hold 15% of the combined value — read it against the metric below, since a higher number is not "better" for every dataset.

Creators in this ranking

5 creators appear in this window; the three most-represented hold 80% of the entries.

CreatorEntriesHighest valueTop entry
NVIDIA783.3Nemotron 3 Ultra 550B A55B (Reasoning)
Allen Institute for AI588.9Olmo 3.1 32B Think
MBZUAI Institute of Foundation Models488.9K2 Think V2
Swiss AI Initiative288.9Apertus 70B Instruct
OpenBMB283.3MiniCPM5-1B (Reasoning)

Where these models appear in other rankings

The same models are measured across multiple independent datasets. Aggregating the 20 entries above against every other AI Battle dataset gives this cross-section, computed by AI Battle and published nowhere upstream:

ModelDatasetsBest placements elsewhere
K2 Think V22#1/20 in Artificial Analysis Openness Index: Components
K2-V2 (high)2#2/20 in Artificial Analysis Openness Index: Components
K2-V2 (medium)2#3/20 in Artificial Analysis Openness Index: Components
K2-V2 (low)2#4/20 in Artificial Analysis Openness Index: Components
Olmo 3.1 32B Think2#5/20 in Artificial Analysis Openness Index: Components
Olmo 3.1 32B Instruct2#6/20 in Artificial Analysis Openness Index: Components
Olmo 3 7B Think2#7/20 in Artificial Analysis Openness Index: Components
Molmo 7B-D2#8/20 in Artificial Analysis Openness Index: Components
Olmo 3 7B Instruct2#9/20 in Artificial Analysis Openness Index: Components
Apertus 70B Instruct2#10/20 in Artificial Analysis Openness Index: Components
Apertus 8B Instruct2#11/20 in Artificial Analysis Openness Index: Components
Nemotron 3 Ultra 550B A55B (Reasoning)45#1/12 in Artificial Analysis Openness Index: Components, #2/20 in 𝜏³-Banking: Time per Task, #3/20 in Time per Intelligence Index Task
Nemotron 3.5 Lightning2#13/20 in Artificial Analysis Openness Index: Components
Nemotron 3 Super 120B A12B (Reasoning)2#14/20 in Artificial Analysis Openness Index: Components
Nemotron Cascade 2 30B A3B2#15/20 in Artificial Analysis Openness Index: Components
NVIDIA Nemotron 3 Nano 30B A3B (Reasoning)2#16/20 in Artificial Analysis Openness Index: Components
MiniCPM5-1B (Reasoning)2#17/20 in Artificial Analysis Openness Index: Components
MiniCPM5-1B (Non-reasoning)2#18/20 in Artificial Analysis Openness Index: Components
NVIDIA Nemotron 3 Nano 4B2#19/20 in Artificial Analysis Openness Index: Components
NVIDIA Nemotron 3 Nano 30B A3B (Non-reasoning)2#20/20 in Artificial Analysis Openness Index: Components

Metric reference

This ranking carries the metric opennessIndex from the Artificial Analysis Openness Index: Score dataset in the tab::eval-openness family. Units and the direction of "better" are not declared by the source dataset, so AI Battle does not invent them: values are shown exactly as published, and the ordering is the source dataset's own.

How this ranking is built

Rebuilt on from the public dataset Artificial Analysis Openness Index: Score (family tab::eval-openness). The source lists 20 entries and all 20 are shown here. AI Battle requests the dataset's rows, keeps the ordering the source publishes, and computes only the statistics on this page: median, mean, spread, the #1–#2 gap, the creator breakdown and the raw-data file. Nothing on this page is generated text about the entries — the numbers come from the source.

Refresh cadence: every deploy re-pulls the dataset, so this page tracks the source's own update cycle. Corrections: see the methodology page.

Limitations

Raw data

Machine-readable copy of this table, with the statistics below and the source provenance: /data/rankings/eval-openness-artificial-analysis-openness-index-score.json. Free to reuse with attribution to AI Battle and the source dataset.

FAQ — Artificial Analysis Openness Index: Score

What is the best model for Artificial Analysis Openness Index: Score in 2026?

K2 Think V2 by MBZUAI Institute of Foundation Models currently leads the Artificial Analysis Openness Index: Score ranking with a score of 88.9, the highest value in this dataset across eval openness. AI Battle aggregates the dataset from its public source and re-checks it daily.

How often is the Artificial Analysis Openness Index: Score ranking updated?

The Artificial Analysis Openness Index: Score ranking is rebuilt daily from the upstream public dataset. Scores are the source dataset's own published values, so positions track whatever that source last released — each ranking page names the dataset it comes from.

Which models rank in the top 3 for Artificial Analysis Openness Index: Score?

The top 3 are: 1. K2 Think V2 (88.9), 2. K2-V2 (high) (88.9), 3. K2-V2 (medium) (88.9).

All AI Battle Rankings

Directories