Skip to content
AI Atlas
ModelActiveOpen weights

AI21-Jamba-Reasoning-3B

AI21 Labsfamily · Jambahuggingface.co/ai21labs/AI21-Jamba-Reasoning-3B

Open in Graph
data quality53

Updated 5 h ago · first seen 11 Sept 2026

model_01M294WVH31A6V1D0JAERSX5WX

Overview

Identity

Canonical model
Yesidentity confidence: highOne 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
1 quantization0 official · 1 third-partySeparate 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

Open weights— weights downloadable under Apache-2.0; commercial use allowed; redistribution allowed; derivatives allowed; 4 dimensions unknown.

Weights downloadable under a permissive or Creative Commons licence allowing commercial use; code or data may be missing.

  • Weights

    Yes

  • Inference code

    —

  • Training code

    —

  • Training data

    —

  • Dataset

    —

  • Commercial use

    Yes

  • Redistribution

    Yes

  • Derivatives

    Yes

Licence: Apache License 2.0 (permissive · SPDX Apache-2.0 · stated as “apache-2.0”)

dimensions marked null are unknown, not false

Key facts

Release date

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Architecture

Architecture

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Model type

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Parameters

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Weights dtype

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

File size

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Library name

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Pipeline tag

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Hugging Face repo

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 15 h agomedium

Hardware fit37

Estimated

37 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit—2.3 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit—2.3 GB est.Yes
Apple M2 Ultra4bit—2.3 GB est.Yes
Apple M1 Ultra4bit—2.3 GB est.Yes
Apple M3 Max4bit—2.3 GB est.Yes
Apple M4 Max4bit—2.3 GB est.Yes
Mac Studio (Apple M5 Max)4bit—2.3 GB est.Yes
MacBook Pro (Apple M5 Max)4bit—2.3 GB est.Yes
Apple M2 Max4bit—2.3 GB est.Yes
Apple M1 Max4bit—2.3 GB est.Yes
Apple M4 Pro4bit—2.3 GB est.Yes
Mac mini (Apple M5 Pro)4bit—2.3 GB est.Yes
MacBook Pro (Apple M5 Pro)4bit—2.3 GB est.Yes
Apple M3 Pro4bit—2.3 GB est.Yes
Apple M1 Pro4bit—2.3 GB est.Yes
Apple M2 Pro4bit—2.3 GB est.Yes
Apple M44bit—2.3 GB est.Yes
iMac (Apple M4)4bit—2.3 GB est.Yes
Mac mini (Apple M6)4bit—2.3 GB est.Yes
MacBook Air (Apple M5)4bit—2.3 GB est.Yes
MacBook Pro (Apple M5)4bit—2.3 GB est.Yes
Apple M24bit—2.3 GB est.Yes
Apple M34bit—2.3 GB est.Yes
Apple M14bit—2.3 GB est.Yes
NVIDIA GeForce RTX 30904bit24 GB2.3 GB est.Yes
NVIDIA GeForce RTX 40904bit24 GB2.3 GB est.Yes
NVIDIA GeForce RTX 50904bit32 GB2.3 GB est.Yes
NVIDIA A100 80GB4bit80 GB2.3 GB est.Yes
NVIDIA H100 SXM4bit80 GB2.3 GB est.Yes
NVIDIA H100 NVL4bit94 GB2.3 GB est.Yes
NVIDIA DGX Spark4bit128 GB2.3 GB est.Yes
NVIDIA H2004bit141 GB2.3 GB est.Yes
NVIDIA H200 NVL4bit141 GB2.3 GB est.Yes
NVIDIA B2004bit180 GB2.3 GB est.Yes
AMD Instinct MI300X4bit192 GB2.3 GB est.Yes
AMD Instinct MI325X4bit256 GB2.3 GB est.Yes
NVIDIA DGX B2004bit1,440 GB2.3 GB est.Yes
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 →
Explicit derived_from / fine_tuned_from / distilled_from relations stated by sources; artifacts collapsed by kind.
DESCENDANTS 0 · ARTIFACTS1 quantizationartifacts · collapsed1 quantization — artifacts · collapsedAI21-Jamba-Reasoning-3B3.2B params · this modelAI21-Jamba-Reasoning-3B — 3.2B params · this model

    Versions & Artifacts1

    Version history

    Licensefirst observation only

    11 Sept 2026current

    Opennessfirst observation only

    11 Sept 2026current

    Parametersfirst 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 1

    quantization 1

    Papers1

    Change history20

    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 Mar 2026 — attributes exactly as the atlas knew them on that day; later corrections are not shown.

    Back to today →

    AI21-Jamba-Reasoning-3B was not yet in AI Atlas on 12 Mar 2026

    First seen 11 Sept 2026. Nothing is inferred backwards: no attribute is shown for dates before the first observation.
    20 claims · 20 properties

    Release daterelease_date1

    Claim history for Release date
    ValueValid from → toStatusSourceConfidenceExtractor
    5 Oct 2025 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Opennessopenness1

    Claim history for Openness
    ValueValid from → toStatusSourceConfidenceExtractor
    open-weights → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Licenselicense1

    Claim history for License
    ValueValid from → toStatusSourceConfidenceExtractor
    apache-2.0 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Architecturearchitecture1

    Claim history for Architecture
    ValueValid from → toStatusSourceConfidenceExtractor
    JambaForCausalLM → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Parametersparameter_count1

    Claim history for Parameters
    ValueValid from → toStatusSourceConfidenceExtractor
    3.2B → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    File sizefile_size_gb1

    Claim history for File size
    ValueValid from → toStatusSourceConfidenceExtractor
    6.4 GB → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Hugging Face repohf_repo1

    Claim history for Hugging Face repo
    ValueValid from → toStatusSourceConfidenceExtractor
    ai21labs/AI21-Jamba-Reasoning-3B → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Pipeline tagpipeline_tag1

    Claim history for Pipeline tag
    ValueValid from → toStatusSourceConfidenceExtractor
    text-generation → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Model cardmodel_card_url1

    Claim history for Model card
    ValueValid from → toStatusSourceConfidenceExtractor
    https://huggingface.co/ai21labs/AI21-Jamba-Reasoning-3B → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Downloadsmetric.downloads1

    Claim history for Downloads
    ValueValid from → toStatusSourceConfidenceExtractor
    17,436 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Likesmetric.likes1

    Claim history for Likes
    ValueValid from → toStatusSourceConfidenceExtractor
    140 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Gatedgated1

    Claim history for Gated
    ValueValid from → toStatusSourceConfidenceExtractor
    No → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Last modifiedlast_modified1

    Claim history for Last modified
    ValueValid from → toStatusSourceConfidenceExtractor
    2025-10-08T13:06:09+00:00 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Library namelibrary_name1

    Claim history for Library name
    ValueValid from → toStatusSourceConfidenceExtractor
    transformers → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    License namelicense_name1

    Claim history for License name
    ValueValid from → toStatusSourceConfidenceExtractor
    jamba-open-model-license → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    License urllicense_url1

    Claim history for License url
    ValueValid from → toStatusSourceConfidenceExtractor
    https://www.ai21.com/licenses/jamba-open-model-license → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Downloads all timemetric.downloads_all_time1

    Claim history for Downloads all time
    ValueValid from → toStatusSourceConfidenceExtractor
    54,672 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Model typemodel_type1

    Claim history for Model type
    ValueValid from → toStatusSourceConfidenceExtractor
    jamba → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Tagstags1

    Claim history for Tags
    ValueValid from → toStatusSourceConfidenceExtractor
    transformers, safetensors, jamba, text-generation → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Weights dtypeweights_dtype1

    Claim history for Weights dtype
    ValueValid from → toStatusSourceConfidenceExtractor
    BF16 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    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

    26

    Source tiers

    T226

    Freshest observation

    5 h ago

    Conflicts

    None

    Source documents 3

    Source documents
    SourceDocumentTypeTierLast observedSnapshots
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/models?author=ai21labs&p=0&sort=downloads listingT2· Quality secondary5 h ago7
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/ai21labs/AI21-Jamba-Reasoning-3B/raw/main/README.md model_cardT2· Quality secondary8 h ago1
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/ai21labs/AI21-Jamba-Reasoning-3B model_pageT2· Quality secondary9 h ago6

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

    Data quality (53/100) measures how well AI Atlas knows this entity — completeness, primary-source ratio, freshness, conflicts — never how good the model is. Methodology →