tiny-aya-global
Coherefamily · Ayadocs.cohere.com/docs/models
Tiny Aya Global is a 3.35B instruction-tuned multilingual model with the best balance across languages and regions. Supports 70 languages.
Updated 3 h ago · first seen 11 Sept 2026
model_01M294H9A6MD8XR7FS8WAHSK2R
Overview
Identity
- Canonical model
- Yesidentity confidence: mediumOne row per real model release. Artifacts (checkpoints, quantisations, conversions) and folded evaluation variants point here.
- Official checkpoints
- official_checkpoints = hf_repo identifiers carried by the model itself; artifacts are separate entities pointing here through canonical_id.
- Artifacts
- None recordedSeparate entities (checkpoint · quantization · conversion · packaging) pointing to this model through canonical_id.
- Provider deployments
- None recorded
- API aliases
- tiny-aya-globalIdentifiers under which providers and evaluators refer to this model.
- Folded evaluation variants
- 0Effort / thinking variants (…-high, …-non-reasoning) are result configurations of this model, not separate models. Their old URLs redirect here.
Openness
Restricted weights— weights downloadable under CC-BY-NC-4.0; commercial use restricted; redistribution allowed; derivatives allowed; 4 dimensions unknown.
Weights downloadable, but the licence restricts commercial use, hosting, derivatives or field of use (community, research and RAIL licences).
Weights
Yes
Inference code
—
Training code
—
Training data
—
Dataset
—
Commercial use
No
Redistribution
Yes
Derivatives
Yes
Licence: Creative Commons Attribution-NonCommercial 4.0 (creative-commons · SPDX CC-BY-NC-4.0 · stated as “cc-by-nc-4.0”)
dimensions marked null are unknown, not false
Key facts
- Release date
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Status
Source:Cohere — docs & blogT1observed 13 h agohigh
- Official page
Source:Cohere — docs & blogT1observed 13 h agohigh
- Model card
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- API model id
Source:Cohere — docs & blogT1observed 13 h agohigh
Architecture
- Architecture
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Model type
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Parameters
Source:Cohere — docs & blogT1observed 13 h agohigh
- Weights dtype
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- File size
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Library name
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Pipeline tag
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
- Hugging Face repo
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
Capabilities
Modalities
- Modalities
- text
- Input
- text
Capabilities
Tool calling
Unavailable
Structured output
Unavailable
Reasoning
No
Artificial Analysis · T2
Vision
No
Cohere — docs & blog · T1
Audio
Unavailable
Fine-tuning available
Unavailable
- Context window
Source:Cohere — docs & blogT1observed 13 h agohigh
- Max output
Source:Cohere — docs & blogT1observed 13 h agohigh
- Languages
Source:Hugging Face Hub (public pages, model cards, papers)T2observed 13 h agomedium
Benchmarks12
Compare with another model →Comparable same task and conditions · Partially comparable same task, conditions differ (effort, temperature, judge) · Not comparable different variant or metric
Current rows only, grouped by benchmark → canonical metric → comparability group (task configuration). Effort variants folded into this model appear as rows of the same group. 12 current rows in total. “vs leader” compares with the current leader of the benchmark's primary group only; other groups are not directly comparable. Comparability rules →
Hardware fit37
Assumptions (7)
- Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead (or the observed artifact file size when one is recorded).
- bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
- KV cache: 2 × layers × kv_heads × head_dim × 2 bytes × context × batch when the architecture is known; otherwise 0.5 GB per 8 192 tokens (× batch), independent of architecture (GQA/MLA models need less).
- A model 'fits' when the estimate is at most the device memory minus 2 GB reserved for the OS and framework.
- Mixture-of-experts models are estimated on total parameters (all experts must be resident); active parameters are ignored.
- Device memory uses the largest configuration when several are listed (e.g. Apple silicon tiers).
- Multi-GPU: device memories are summed; interconnect bandwidth, tensor-parallel replication and pipeline bubbles are not modelled.
Lineage
Open in Graph →- ancestor: tiny-aya-base
Versions & Artifacts0
Version history
Openness2 changes
11 Sept 2026→11 Sept 2026→11 Sept 2026current
Context windowfirst observation only
11 Sept 2026current
Licensefirst observation only
11 Sept 2026current
Max outputfirst observation only
11 Sept 2026current
Parametersfirst observation only
11 Sept 2026current
Statusfirst observation only
11 Sept 2026current
Each hop is a claim: click a value for its source, tier and observation time. Nothing is overwritten — a new observation closes the previous claim.
Artifacts 0
No artifact (checkpoint, quantisation, conversion or packaging) points to this model yet.
Papers1
- arXiv:2603.11510Active35
Timeline7
Full timeline →tiny-aya-global scores 0% on Terminal-Bench
artificial_analysistiny-aya-global scores 20.14% on IFBench
artificial_analysistiny-aya-global scores 5.24% on Humanity's Last Exam
artificial_analysistiny-aya-global scores 30.51% on GPQA Diamond
artificial_analysistiny-aya-global scores 4.83 on Artificial Analysis Intelligence Index
artificial_analysistiny-aya-global: openness changed from open-weights to restricted
Opennessopen-weights→restrictedhuggingface
Change history52
Viewing AI Atlas as of 5 Sept 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.
Back to today →tiny-aya-global was not yet in AI Atlas on 5 Sept 2026
Familyfamily1
Release daterelease_date1
Statusstatus1
Opennessopenness3
Licenselicense1
Architecturearchitecture1
Parametersparameter_count11conflicting claims
Context windowcontext_length2conflicting claims
Max outputmax_output_tokens1
Modalitiesmodalities1
Input modalitiesmodalities_input1
Languageslanguages1
API model idapi_model_id1
Base modelbase_model1
File sizefile_size_gb1
Hugging Face repohf_repo1
Pipeline tagpipeline_tag1
Official pageofficial_url1
Model cardmodel_card_url1
Downloadsmetric.downloads1
Likesmetric.likes1
Aa context windowaa_context_window1
Aa opennessaa_openness1
Commercial use allowedcommercial_use_allowed1
Derivatives allowedderivatives_allowed1
Descriptiondescription1
Endpointsendpoints1
Gatedgated1
Hf inference providershf_inference_providers1
Last modifiedlast_modified1
Library namelibrary_name1
Downloads all timemetric.downloads_all_time1
Model typemodel_type1
Reasoningreasoning1
Redistribution allowedredistribution_allowed1
Visionvision1
Weights availableweights_available1
Weights dtypeweights_dtype1
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
Provenance
Attributed facts
40
Source tiers
T1T212 / 28
Freshest observation
3 h ago
Conflicts
11 flagged
Source documents 4
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.
Data quality (78/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →