Skip to main content

Cohere LabsLanguage modelOpen weights

Tiny Aya Earth

Tiny Aya Earth has no overall index value yet because there are too few independent benchmark results.

PopularityRank 291 of 342 language models

21out of 100new calculation

based on two data groups, mostly search, range 11 to 31

Summary

Tiny Aya Earth focuses on languages across Africa and West Asia. Cohere Labs offers it as a regional alternative to the balanced Global variant. The checkpoint has 3.35 billion parameters and belongs to the Tiny Aya family covering more than 70 languages. Its local weights use CC BY-NC 4.0 with additional Cohere terms for non-commercial use.

All benchmark results

Every published result with its source, test setup and comparison models.

Show all 66 published benchmark results

Page 1 of 6

Measurements without matching peer values

These measurements have no matching peer values under the same test conditions. Their original values and sources remain available here.

NormAd English Stories, Europe accuracy (Show measurement, test conditions, and source)
Source value
74.22
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd English Stories, West Asia accuracy (Show measurement, test conditions, and source)
Source value
68.09
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd English Stories, South Asia accuracy (Show measurement, test conditions, and source)
Source value
68.37
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd English Stories, Asia-Pacific accuracy (Show measurement, test conditions, and source)
Source value
77.23
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd English Stories, Africa accuracy (Show measurement, test conditions, and source)
Source value
73.14
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd English Stories, Americas accuracy (Show measurement, test conditions, and source)
Source value
72.63
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 16, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 16; NormAd per-region accuracy. Stories are in English. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, Europe accuracy (Show measurement, test conditions, and source)
Source value
71.98
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, West Asia accuracy (Show measurement, test conditions, and source)
Source value
66.35
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, South Asia accuracy (Show measurement, test conditions, and source)
Source value
70.47
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, Asia-Pacific accuracy (Show measurement, test conditions, and source)
Source value
75.4
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, Africa accuracy (Show measurement, test conditions, and source)
Source value
66.19
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
NormAd Source-Language Stories, Americas accuracy (Show measurement, test conditions, and source)
Source value
70.71
Metric
accuracy
Unit
%
Benchmark version
NormAd, Table 17, arXiv:2603.11510v1
Direction
Higher is better
Source type
vendor-reported
Evaluator
Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1
Status
active
Retrieved at
2026-10-17
Methodology
Vendor-reported result. Exact Tiny Aya Earth row from Table 17; NormAd per-region accuracy. Stories are in each country's official language. The paper maps 75 countries to languages but does not disclose this row's scored-example denominator or prompt and decoding details.
Source Cohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1

Page 1 of 6

Popularity compared

How often the model is searched for, used and mentioned, compared with every model of its category.

Popularity index

The 100 most popular models of this category, 0 to 100, higher is more popular

20 of 101 models

As of November 2026 · Source: gradually.ai popularity index

Over time

Weekly values from October 18, 2026 to November 20, 2026

Tiny Aya EarthIndex-Homura-2B 21Index-Homura-9B 21Index-NativeLong-2B 21Index-NativeLong-9B 21

Method changes in the timeline, a jump at these points is no market event:

  1. 1November 1, 2026: Value only with search or usage data, 30-day attention
  2. 2November 1, 2026: Method version 4 (common scale, shrinkage)
  3. 3November 9, 2026: New topics measurement V20 (daily quorum, short forms)

What is gaining popularity right now, across models, vendors, harnesses and AI subscriptions, is collected in All AI trends.

All specifications

Architecture

Parameters
3.35 billion
Source: Parameters

Input and output

Languages
More than 70 languages across the Tiny Aya family
Source: Languages
Context window
8K input tokens (model card)
Source: Context window

Watermarking and provenance

AI watermarking refers to invisible signals embedded in the output. Visible marks and provenance metadata are listed separately. Unknown means there is insufficient evidence for that particular field.

AI watermarking
Unknown
no source
Watermarking technique
Not known
no source

Licence and access

Release
February 17, 2026
Source: Release
License
CC BY-NC 4.0 with Cohere usage terms
Source: License
Shutdown
None announced
no source

Further details

Specialization
Strengths for languages across Africa and West Asia
Source: Specialization
Weight download
After accepting Cohere’s terms on Hugging Face
Source: Weight download

Frequently asked questions about Tiny Aya Earth

Can I run Tiny Aya Earth locally?

Yes, the weights are open.

Stated by Cohere Labs

Open weights
yes
License
CC BY-NC 4.0 with Cohere usage terms
Parameters
3.4B
active per token
n/a

Sources and data date

Show all 18 sources

Benchmark sources at a glance

Which independent evaluators tested this model. This is coverage, not a quality ranking.

No active result from the reviewed external evaluators can currently be attributed to this model. Other source evidence remains available below.

Without vendor claims and historical or disputed results; those stay labelled in the single values.

TypeEvidence and data date
Research statusResearch date October 4, 2026. 10 source URLs checked. This documents the inspected sources, not an exhaustive inventory of every publication.
Not published

Grade-stratified confidence intervals

Original wording

The inspected method/result sources report confidence intervals for overall accuracy only, not the grade-specific cells.

Unresolved

Overall item-count reconciliation

Original wording

The method article reports 2,569 items while the pinned results file reports 2,570 overall items; the difference is not explained.

Additional sourceCohere Labs Tiny Aya announcement (retrieved October 3, 2026; October 4, 2026) · Cohere Labs Tiny Aya announcement · Editorial description reviewed October 4, 2026
Additional sourceCohere Labs Aya research page (retrieved October 3, 2026) · Cohere Labs Aya research page · Editorial description reviewed October 4, 2026
Additional sourceTiny Aya Earth model card (retrieved October 3, 2026) · Tiny Aya Earth model card · Editorial description reviewed October 4, 2026
Additional sourceCohere Labs, Tiny Aya: Bridging Scale and Multilingual Depth, arXiv:2603.11510v1 (retrieved October 17, 2026)
Additional sourceSwahili Understanding Benchmark v1.0 results (retrieved October 17, 2026; October 18, 2026)
Additional sourceCohere Labs and Cohere, Tiny Aya: Bridging Scale and Multilingual Depth (arXiv:2603.11510) (retrieved October 23, 2026)
Additional sourceModel metadata (retrieved October 17, 2026)123456
Additional sourceOpenRouter: Popularity index (derived), CC BY 4.0
Additional sourceVercel AI Gateway: Popularity index (derived), CC BY 4.0
Additional sourceOllama: cumulative downloads for the popularity index
Additional sourceReplicate: cumulative runs for the popularity index
Additional sourceGoogle Trending Now: limited feed selection for Germany and the US
Additional sourceCreative Commons Attribution 4.0