Skip to content
AI Atlas
ModelAnnouncedRestricted weights

North Small Translate

Coherefamily · North

Open in Graph
data quality62

Updated 42 min ago · first seen 11 Sept 2026

model_01M29DV4DKDBPPS82CHKF70QR4

Overview

Identity

Canonical model
Yesidentity confidence: highOne row per real model release. Artifacts (checkpoints, quantisations, conversions) and folded evaluation variants point here.
Official checkpoints
None recordedofficial_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
NoneIdentifiers 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:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Status

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Version

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Architecture

Architecture

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Parameters

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Active parameters

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Mixture of experts

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Capabilities

Modalities

Modalities
text
Input
text
Output
text

Capabilities

  • Tool calling

    Unavailable

  • Structured output

    Unavailable

  • Reasoning

    Unavailable

  • Vision

    Unavailable

  • Audio

    Unavailable

  • Fine-tuning available

    Unavailable

No capability flags have been observed from a source yet — we do not infer them.

Context window

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Max output

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Languages

Source:Cohere — docs & blogT2observed 8 h agomediumLLM-extracted

Hardware fit37

Estimated

15 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit—125.8 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit—125.8 GB est.Yes
Apple M2 Ultra4bit—125.8 GB est.Yes
Apple M1 Ultra4bit—125.8 GB est.Yes
Apple M3 Max4bit—125.8 GB est.Yes
Apple M4 Max4bit—125.8 GB est.Yes
Mac Studio (Apple M5 Max)4bit—125.8 GB est.Yes
MacBook Pro (Apple M5 Max)4bit—125.8 GB est.Yes
NVIDIA DGX Spark4bit128 GB125.8 GB est.Yes
NVIDIA H2004bit141 GB125.8 GB est.Yes
NVIDIA H200 NVL4bit141 GB125.8 GB est.Yes
NVIDIA B2004bit180 GB125.8 GB est.Yes
AMD Instinct MI300X4bit192 GB125.8 GB est.Yes
AMD Instinct MI325X4bit256 GB125.8 GB est.Yes
NVIDIA DGX B2004bit1,440 GB125.8 GB est.Yes
Apple M2 Max4bit—125.8 GB est.No
Apple M1 Max4bit—125.8 GB est.No
Apple M4 Pro4bit—125.8 GB est.No
Mac mini (Apple M5 Pro)4bit—125.8 GB est.No
MacBook Pro (Apple M5 Pro)4bit—125.8 GB est.No
Apple M3 Pro4bit—125.8 GB est.No
Apple M1 Pro4bit—125.8 GB est.No
Apple M2 Pro4bit—125.8 GB est.No
Apple M44bit—125.8 GB est.No
iMac (Apple M4)4bit—125.8 GB est.No
Mac mini (Apple M6)4bit—125.8 GB est.No
MacBook Air (Apple M5)4bit—125.8 GB est.No
MacBook Pro (Apple M5)4bit—125.8 GB est.No
Apple M24bit—125.8 GB est.No
Apple M34bit—125.8 GB est.No
Apple M14bit—125.8 GB est.No
NVIDIA GeForce RTX 30904bit24 GB125.8 GB est.No
NVIDIA GeForce RTX 40904bit24 GB125.8 GB est.No
NVIDIA GeForce RTX 50904bit32 GB125.8 GB est.No
NVIDIA A100 80GB4bit80 GB125.8 GB est.No
NVIDIA H100 SXM4bit80 GB125.8 GB est.No
NVIDIA H100 NVL4bit94 GB125.8 GB est.No
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.

Versions & Artifacts0

Version history

Openness3 changes

11 Sept 2026→12 Sept 2026→12 Sept 2026→12 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.

Timeline2

Full timeline →

Change history17

Temporal, append-only claims: a new observation closes the previous claim instead of overwriting it. Rewind the record with the as-of picker.

Viewing AI Atlas as of 12 Sept 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

Back to today →

Attributes as of 12 Sept 2026 22 claims in force

Family
North
Version
1.0
Release date
10 Sept 2026
Status
announced
Openness
restricted-weights
License
CC-BY-NC-4.0
Architecture
MoE
Parameters
218B
Active parameters
25B
Mixture of experts
Yes
Context window
16K tokens
Max output
16K tokens
Modalities
text
Input modalities
text
Output modalities
text
Languages
50+ languages
Commercial use allowed
No
Derivatives allowed
Yes
License key
CC-BY-NC-4.0
Redistribution allowed
Yes
Weights available
Yes
17 claims · 17 properties

Familyfamily1

Claim history for Family
ValueValid from → toStatusSourceConfidenceExtractor
North → currentcurrentCohere — docs & blogT2mediumllm

Versionversion1

Claim history for Version
ValueValid from → toStatusSourceConfidenceExtractor
1.0 → currentcurrentCohere — docs & blogT2mediumllm

Release daterelease_date1

Claim history for Release date
ValueValid from → toStatusSourceConfidenceExtractor
10 Sept 2026 → currentcurrentCohere — docs & blogT2mediumllm

Statusstatus1

Claim history for Status
ValueValid from → toStatusSourceConfidenceExtractor
announced → currentcurrentCohere — docs & blogT2mediumllm

Opennessopenness1

Claim history for Openness
ValueValid from → toStatusSourceConfidenceExtractor
open-weights → currentcurrentCohere — docs & blogT2mediumllm

Licenselicense1

Claim history for License
ValueValid from → toStatusSourceConfidenceExtractor
CC BY-NC 4.0 → currentcurrentCohere — docs & blogT2mediumllm

Architecturearchitecture1

Claim history for Architecture
ValueValid from → toStatusSourceConfidenceExtractor
MoE → currentcurrentCohere — docs & blogT2mediumllm

Parametersparameter_count1

Claim history for Parameters
ValueValid from → toStatusSourceConfidenceExtractor
218B → currentcurrentCohere — docs & blogT2mediumllm

Active parametersactive_parameter_count1

Claim history for Active parameters
ValueValid from → toStatusSourceConfidenceExtractor
25B → currentcurrentCohere — docs & blogT2mediumllm

Mixture of expertsis_moe1

Claim history for Mixture of experts
ValueValid from → toStatusSourceConfidenceExtractor
Yes → currentcurrentCohere — docs & blogT2mediumllm

Context windowcontext_length1

Claim history for Context window
ValueValid from → toStatusSourceConfidenceExtractor
16K tokens → currentcurrentCohere — docs & blogT2mediumllm

Max outputmax_output_tokens1

Claim history for Max output
ValueValid from → toStatusSourceConfidenceExtractor
16K tokens → currentcurrentCohere — docs & blogT2mediumllm

Modalitiesmodalities1

Claim history for Modalities
ValueValid from → toStatusSourceConfidenceExtractor
Text → currentcurrentCohere — docs & blogT2mediumllm

Input modalitiesmodalities_input1

Claim history for Input modalities
ValueValid from → toStatusSourceConfidenceExtractor
Text → currentcurrentCohere — docs & blogT2mediumllm

Output modalitiesmodalities_output1

Claim history for Output modalities
ValueValid from → toStatusSourceConfidenceExtractor
Text → currentcurrentCohere — docs & blogT2mediumllm

Languageslanguages1

Claim history for Languages
ValueValid from → toStatusSourceConfidenceExtractor
50+ languages → currentcurrentCohere — docs & blogT2mediumllm

Hardware requirementshardware_requirements1

Claim history for Hardware requirements
ValueValid from → toStatusSourceConfidenceExtractor
1× B200 @ W4A4 or 2× H100s @ W4A4 → currentcurrentCohere — docs & blogT2mediumllm

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

22

Source tiers

T222

Freshest observation

42 min ago

Conflicts

None

Source documents 1

Source documents
SourceDocumentTypeTierLast observedSnapshots
Cohere — docs & blogcohere.com/blog/north-small-translate newsT1· Official56 min ago8

Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.

Data quality (62/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →