Skip to content
AI Atlas
ModelActive

MiniMax-M3

MiniMaxhuggingface.co/MiniMaxAI/MiniMax-M3

Open in Graph
data quality53

Updated 6 h ago · first seen 11 Sept 2026

model_01M294Z7RP183FSMR9D4AWSDS7

Overview

Identity

Identity block not returned by the API for this entity.

Openness

Openness not classified yet — no sourced evidence to place this model in the ontology.

Key facts

Release date

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

Model card
huggingface.co/MiniMaxAI/MiniMax-M3

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Architecture

Architecture

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

Model type

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

Parameters

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Weights dtype

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

File size

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

Library name

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 11 h agomedium

Pipeline tag

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Hugging Face repo

Source:Hugging Face Hub (public pages, model cards, papers)T2observed 12 h agomedium

Hardware fit37

Estimated

4 of 37 device × quantization combinations fit.

Run locally: your machine →
Estimated hardware fit
HardwareQuantizationDevice memoryEst. memoryFits
Apple M3 Ultra4bit—246.1 GB est.Yes
Mac Studio (Apple M5 Ultra)4bit—246.1 GB est.Yes
AMD Instinct MI325X4bit256 GB246.1 GB est.Yes
NVIDIA DGX B2004bit1,440 GB246.1 GB est.Yes
Apple M2 Ultra4bit—246.1 GB est.No
Apple M1 Ultra4bit—246.1 GB est.No
Apple M3 Max4bit—246.1 GB est.No
Apple M4 Max4bit—246.1 GB est.No
Mac Studio (Apple M5 Max)4bit—246.1 GB est.No
MacBook Pro (Apple M5 Max)4bit—246.1 GB est.No
Apple M2 Max4bit—246.1 GB est.No
Apple M1 Max4bit—246.1 GB est.No
Apple M4 Pro4bit—246.1 GB est.No
Mac mini (Apple M5 Pro)4bit—246.1 GB est.No
MacBook Pro (Apple M5 Pro)4bit—246.1 GB est.No
Apple M3 Pro4bit—246.1 GB est.No
Apple M1 Pro4bit—246.1 GB est.No
Apple M2 Pro4bit—246.1 GB est.No
Apple M44bit—246.1 GB est.No
iMac (Apple M4)4bit—246.1 GB est.No
Mac mini (Apple M6)4bit—246.1 GB est.No
MacBook Air (Apple M5)4bit—246.1 GB est.No
MacBook Pro (Apple M5)4bit—246.1 GB est.No
Apple M24bit—246.1 GB est.No
Apple M34bit—246.1 GB est.No
Apple M14bit—246.1 GB est.No
NVIDIA GeForce RTX 30904bit24 GB246.1 GB est.No
NVIDIA GeForce RTX 40904bit24 GB246.1 GB est.No
NVIDIA GeForce RTX 50904bit32 GB246.1 GB est.No
NVIDIA A100 80GB4bit80 GB246.1 GB est.No
NVIDIA H100 SXM4bit80 GB246.1 GB est.No
NVIDIA H100 NVL4bit94 GB246.1 GB est.No
NVIDIA DGX Spark4bit128 GB246.1 GB est.No
NVIDIA H2004bit141 GB246.1 GB est.No
NVIDIA H200 NVL4bit141 GB246.1 GB est.No
NVIDIA B2004bit180 GB246.1 GB est.No
AMD Instinct MI300X4bit192 GB246.1 GB est.No
Assumptions (6)
  • Estimated, not measured: weights = parameters × bytes/param × 1.15 runtime overhead.
  • bytes/param: 4bit = 0.5, 8bit = 1.0, fp16 = 2.0 (uniform quantization, no per-layer exceptions).
  • KV cache approximated at 0.5 GB per 8 192 tokens of context, 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).

Lineage

Open in Graph →
Explicit derived_from / fine_tuned_from / distilled_from relations stated by sources; artifacts collapsed by kind.
DESCENDANTS 0 · ARTIFACTS2 quantizationsartifacts · collapsed2 quantizations — artifacts · collapsedMiniMax-M3427B params · this modelMiniMax-M3 — 427B params · this model

    Papers1

    Timeline1

    Full timeline →

    Change history21

    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 38 claims in force

    Release date
    2 Jun 2026
    Openness
    open-weights
    License
    Other
    Architecture
    MiniMaxM3SparseForConditionalGeneration
    Parameters
    427B
    Context window
    1.05M tokens
    Max output
    512K tokens
    Modalities
    image, text, video
    Input modalities
    image, text, video
    Output modalities
    text
    File size
    854.2 GB
    Hugging Face repo
    MiniMaxAI/Minimax-M3
    Pipeline tag
    image-text-to-text
    Downloads
    186,082
    Likes
    1,530
    Aa context window
    1,000,000
    Aa openness
    open-weights
    Gated
    No
    Hf inference providers
    deepinfra, featherless-ai, fireworks-ai, novita, together
    Last modified
    2026-07-23T04:25:20+00:00
    Library name
    transformers
    License name
    minimax-community
    License url
    LICENSE
    Aa median output tokens per second
    87.7
    Downloads all time
    641,639
    Model type
    minimax_m3_vl
    Openrouter id
    minimax/minimax-m3
    Openrouter listed at
    31 May 2026
    Reasoning
    Yes
    Structured output
    Yes
    Supported parameters
    frequency_penalty, include_reasoning, logit_bias, logprobs, max_tokens, min_p, presence_penalty, reasoning, repetition_penalty, response_format, seed, stop, structured_outputs, temperature, tool_choice, tools, top_k, top_logprobs, top_p
    Tags
    transformers, safetensors, minimax_m3_vl, image-text-to-text, multimodal, moe, agent, coding, video, custom_code
    Tool calling
    Yes
    Vision
    Yes
    Weights available
    Yes
    Weights dtype
    BF16, F32
    21 claims · 21 properties

    Release daterelease_date1

    Claim history for Release date
    ValueValid from → toStatusSourceConfidenceExtractor
    2 Jun 2026 → 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
    other → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Architecturearchitecture1

    Claim history for Architecture
    ValueValid from → toStatusSourceConfidenceExtractor
    MiniMaxM3SparseForConditionalGeneration → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Parametersparameter_count1

    Claim history for Parameters
    ValueValid from → toStatusSourceConfidenceExtractor
    427B → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    File sizefile_size_gb1

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

    Hugging Face repohf_repo1

    Claim history for Hugging Face repo
    ValueValid from → toStatusSourceConfidenceExtractor
    MiniMaxAI/MiniMax-M3 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Pipeline tagpipeline_tag1

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

    Model cardmodel_card_url1

    Claim history for Model card
    ValueValid from → toStatusSourceConfidenceExtractor
    https://huggingface.co/MiniMaxAI/MiniMax-M3 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Downloadsmetric.downloads1

    Claim history for Downloads
    ValueValid from → toStatusSourceConfidenceExtractor
    186,082 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Likesmetric.likes1

    Claim history for Likes
    ValueValid from → toStatusSourceConfidenceExtractor
    1,530 → 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

    Hf inference providershf_inference_providers1

    Claim history for Hf inference providers
    ValueValid from → toStatusSourceConfidenceExtractor
    deepinfra, featherless-ai, fireworks-ai, novita, together → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Last modifiedlast_modified1

    Claim history for Last modified
    ValueValid from → toStatusSourceConfidenceExtractor
    2026-07-23T04:25:20+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
    minimax-community → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    License urllicense_url1

    Claim history for License url
    ValueValid from → toStatusSourceConfidenceExtractor
    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
    641,639 → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Model typemodel_type1

    Claim history for Model type
    ValueValid from → toStatusSourceConfidenceExtractor
    minimax_m3_vl → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Tagstags1

    Claim history for Tags
    ValueValid from → toStatusSourceConfidenceExtractor
    transformers, safetensors, minimax_m3_vl, image-text-to-text, multimodal, moe, agent, coding, video, custom_code → currentcurrentHugging Face Hub (public pages, model cards, papers)T2mediumdeterministic

    Weights dtypeweights_dtype1

    Claim history for Weights dtype
    ValueValid from → toStatusSourceConfidenceExtractor
    BF16, F32 → 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

    21

    Source tiers

    T221

    Freshest observation

    6 h ago

    Conflicts

    None

    Source documents 5

    Source documents
    SourceDocumentTypeTierLast observedSnapshots
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/MiniMaxAI/MiniMax-M3/raw/main/README.md model_cardT2· Quality secondary5 h ago1
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/MiniMaxAI/MiniMax-M3 model_pageT2· Quality secondary6 h ago4
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/MiniMaxAI/MiniMax-M3-MXFP8 model_pageT2· Quality secondary6 h ago5
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/amd/MiniMax-M3-MXFP4 model_pageT2· Quality secondary6 h ago6
    Hugging Face Hub (public pages, model cards, papers)huggingface.co/models?author=MiniMaxAI&p=0&sort=downloads listingT2· Quality secondary6 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 →