reasoning
Motif-3
Motif Technologies 314B-total, 13.2B-active MoE model post-trained for agentic, coding and reasoning tasks, released with open weights.
Levelmodel
Classreasoning
Parameters314B
Context262,144
ReleasedAug 2026
Opennessopen weights permissive
Runs onnot in the notes
Statusactive
Benchmark Results
vendor-reported
AA-LCR v1.1 v1.1long context
temperature 1.0, top-p 0.95, max sequence 262,144; AA-LCRvendor-reported
72.3
050100
Score in %. Higher is better.
Motif-3 model card
CritPtknowledge and reasoning
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
6.6
050100
Score in %. Higher is better.
Motif-3 model card
GPQA Diamondknowledge and reasoning
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
83.4
050100
Score in %. Higher is better.
Motif-3 model card
Humanity's Last Exam (HLE)knowledge and reasoning
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
37
050100
Score in %. Higher is better.
Motif-3 model card
IFBenchknowledge and reasoning
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
78.2
050100
Score in %. Higher is better.
Motif-3 model card
SciCodeknowledge and reasoning
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
40.6
050100
Score in %. Higher is better.
Motif-3 model card
SWE-bench Verifiedcoding
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
76.2
050100
Score in %. Higher is better.
Motif-3 model card
tau2-benchagents and tool use
temperature 1.0, top-p 0.95, max sequence 262,144; Telecomvendor-reported
94.7
050100
Score in %. Higher is better.
Motif-3 model card
Terminal-Bench 2.1agents and tool use
temperature 1.0, top-p 0.95, max sequence 262,144vendor-reported
74.9
050100
Score in %. Higher is better.
Motif-3 model card
Where this comes from
Every score names who ran it: vendor-reported, maintainer-run or independent. A rank is shown with the population it was ranked in (gap rule 3).
Lineage
finetune
Motif-3-Base is the parent (finetune)high confidencemodel cardWhere this comes from
Parent and child relations from model cards, Hugging Face base_model fields and the genealogy tables of the research notes. The tree shows two generations each way, 2 models. Confidence and evidence are the notes' own.
R&D Topics
Identity
- Developer
- Motif Technologies
- Family
- not in the notes
- Released
- Aug 2026 sourcesinferred from page
- Weights published
- –
- Record checked
- 11 Oct 2026
Where this comes from
Developer and dates from the research notes. Release date is the vendor's public announcement; weights date is when open weights were published.
Architecture and Size
- Architecture
- MoE Decoder-only MoE, 53 layers, Grouped Differential Latent Attention, 384 routed experts (top 8) plus 1 shared, MTP head
- Total parameters
- 314B sources
- Active parameters
- 13.2B
- Checkpoint parameters
- –
- Context length
- 262,144 tokens
- Input
- text
- Output
- text
Where this comes from
Parameter counts as the developer states them. Where the checkpoint holds a different total, both are shown (register A16).
Openness and Access
- Openness
- open weights permissive
- Licence
- MIT License sources
- Access routes
- open weights download
Where this comes from
The licence file governs over registry tags and aggregator labels.
Deployment
- Runs on
- not in the notes
- Footprint
- not in the notes
- Runtimes
- not in the notes
Where this comes from
Deployment tiers and runtimes as named in the notes. Footprint is the notes' figure, not a measurement by AURION.
Indonesia
- Indonesian support
- not stated
- Regional languages
- not in the notes
- Jakarta hosting
- unknown
- Languages
- English and Korean language tags
Where this comes from
Indonesia fields are carried on every model; unknown is an allowed value.
Industry Use Cases
No industry use case links this model.
Links
DeveloperMotif Technologiescompany
Motif TechnologiescompanyModelMotif-3Motif Technologies, reasoning
Motif-3Motif Technologies, reasoningR&D TopicsAgents and Tool UseEfficiency Architectures (MoE, Sparse Attention, Long Context)Reasoning via Large-Scale RL
Use CasesProcess Optimisation and Advanced Process ControlIndustrialChinese Medical LLMsHealthcareUse cases of models in the same R&D topic; none is recorded for this model itself.
Who Runs It
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025
Conch CementIn production, 23 Jun 2025
PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.
PT Amman Mineral Nusa Tenggara (AMMAN)In production, 8 Jul 2026MIND IDResearch, 14 Oct 2025Conch CementIn production, 23 Jun 2025PT Kalbe Morinaga IndonesiaIn production, Oct 2024+2 moreDeployments of these use cases; none names this model.Organisations 1
R&D topics 3
Efficiency Architectures (MoE, Sparse Attention, Long Context)Reducing training and inference cost per capability.produced byReasoning via Large-Scale RLReinforcement learning on outcomes, often verifiable math or code, to produce long reason…produced byAgents and Tool UseSystems that plan, call tools and act over multi-step tasks.produced by
Sources
- huggingface.co/Motif-Technologies/Motif-3 https://huggingface.co/Motif-Technologies/Motif-3
To cite this page: Project AURION, AI Use-Cases Research & Innovation, "Motif-3", https://aurion.semai.id/models/motif-technologies-motif-3/, records of October 2026.