# Motif 3 Beta ![Organization](https://img.shields.io/badge/Org-Motif%20Technologies-555) ![Total](https://img.shields.io/badge/Total-314B-1f6feb) ![Active](https://img.shields.io/badge/Active-13B-2da44e) ![Context](https://img.shields.io/badge/Context-256K%20tokens-1f6feb) ![Pricing](https://img.shields.io/badge/Price-Free%20%28Beta%29-2da44e) ![License](https://img.shields.io/badge/License-Non--Commercial-d29922) ![Modality](https://img.shields.io/badge/Modality-Text-8957e5) ![Released](https://img.shields.io/badge/Released-Jul%202026-555) > **Quick answer:** Motif 3 Beta is a 314B-parameter sparse Mixture-of-Experts reasoning model built entirely in-house by Motif Technologies, with ~13B parameters active per token, a 256K-token context window, and a score of 44 on the Artificial Analysis Intelligence Index. Weights are freely downloadable from Hugging Face for non-commercial use; API access is $0.00/MTok during the beta period. ## At a Glance **Where Motif 3 Beta leads** - Fully in-house architecture — GDLA (Grouped Differential Latent Attention), Grouped PolyNorm, and a custom expert routing design, not a derivative of any existing open-source model. - Excellent price-performance: AAII score of 44 with $0.00/MTok pricing places it among the most cost-efficient models at this capability tier. - Long-context native: 256K tokens baked into the base architecture, suitable for multi-document analysis and large codebases. - Ultra-sparse routing: 384 routed experts activating only 8 per token (plus 1 shared), keeping active compute close to a dense 13B model despite 314B total weights. **Where it lags** - Preview / beta checkpoint — not the final release; Motif Technologies explicitly warns the weights may change before the stable release. - Text only — no image, audio, or other input modalities. - Commercial use requires prior written permission from Motif Technologies; not suitable for production deployments without a commercial license. - No dedicated inference service yet; no hosted API beyond the beta model hub. **Best for:** Research and experimentation on long-context reasoning tasks, non-commercial fine-tuning, and evaluation of fully in-house MoE architectures. ## What Motif 3 Beta Is Motif 3 Beta is the first publicly released checkpoint from Motif Technologies' Motif-3 model series, dropped onto Hugging Face in July 2026 as an intermediate preview ahead of the full stable release. The model is noteworthy for being a ground-up proprietary design: Motif Technologies has not re-parameterised or distilled any existing open-source architecture. Instead, it introduces several novel components — Grouped Differential Latent Attention (GDLA), Grouped PolyNorm activation applied per expert, and a customised multi-head compression module (mHC) — all developed internally. Under the hood, Motif 3 uses a sparse MoE decoder stack with 53 layers, 384 routed experts per layer (top-8 selected per token) and a single shared expert, giving a total parameter count of ~314B with only ~13B active per forward pass. The hidden size is 4096 and the vocabulary spans 220,160 tokens, reflecting multilingual training. Weights are stored in bfloat16. The architecture was designed natively for 256K context (262,144 tokens) without additional tricks such as YaRN or ALiBi, making the full context window reliably usable at inference time. On the Artificial Analysis Intelligence Index v4.1, Motif 3 Beta scores 44, placing it at #35 out of 579 evaluated models and well above the median for free-tier ($0.00/MTok) models. Artificial Analysis classifies it as a reasoning model: it uses extended thinking or chain-of-thought reasoning internally before generating a final answer. The model is multilingual with particular emphasis on English and Korean. ## Specifications | Field | Value | |---|---| | Organization | Motif Technologies | | Total parameters | ~314B | | Active parameters | ~13B per token | | Architecture | Sparse MoE — GDLA + Grouped PolyNorm + modified mHC | | Layers | 53 | | Routed experts | 384 (top-8 activated per token) | | Shared experts | 1 | | Hidden size | 4096 | | Vocabulary | 220,160 tokens | | Context window | 262,144 tokens (256K) | | Max output | Not disclosed | | Numerics | BF16 | | Reasoning | Yes (chain-of-thought) | | Input modalities | Text | | Output modalities | Text | | License | Open, non-commercial only (written permission required for commercial use) | | Release date | July 20, 2026 (beta checkpoint; stable release forthcoming) | | HuggingFace | [Motif-Technologies/Motif-3-Beta](https://huggingface.co/Motif-Technologies/Motif-3-Beta) | ## Pricing | | Input (per 1M tokens) | Output (per 1M tokens) | |---|---|---| | Motif Technologies (Beta) | $0.00 | $0.00 | Motif 3 Beta is currently available at no cost via the [Motif model hub](https://model-hub.motiftech.io/en/). Weights are also freely downloadable from Hugging Face for self-hosting. Commercial use requires a separate written agreement with Motif Technologies. ## Context Window Motif 3 Beta has a **256K-token context window** (262,144 tokens) — roughly **≈ 393 A4 pages** of 12pt Arial text in a single request. The long context is native to the base architecture rather than a post-training extension, supporting whole-codebase analysis, multi-document reasoning, and long-conversation tracking without retrieval workarounds. ## Public Benchmark Scores | Benchmark | Score | Source | Date | |---|---|---|---| | [Artificial Analysis Intelligence Index (AAII)](/benchmarks/aa-omniscience) | 44 / 100 | [Artificial Analysis model page](https://artificialanalysis.ai/models/motif-0714) | 2026-07 | *The AAII is a composite index by Artificial Analysis incorporating GDPval-AA v2, τ³-Banking, Terminal-Bench v2.1, SciCode, Humanity's Last Exam, GPQA Diamond, CritPt, AA-Omniscience, and AA-LCR. Scores above are reported by third parties and shown for context; they are not Benchgen measurements. Methodology details at [artificialanalysis.ai/methodology](https://artificialanalysis.ai/methodology/intelligence-benchmarking).* ## Motif 3 Beta vs Alternatives | Model | Context | Parameters (total/active) | AAII | License | Price (in/out per 1M) | |---|---|---|---|---|---| | **Motif 3 Beta** | 256K | 314B / 13B | 44 | Non-commercial | $0.00 / $0.00 | | [Inkling](/models/thinkingmachines/inkling) | 1M | 975B / 41B | — | Apache 2.0 | Free (open weights) | | [Kimi K3](/models/moonshot/kimi-k3) | 1M | 2.8T / — | — | Apache 2.0 (Jul 27) | $3.00 / $15.00 | | [DeepSeek V4 Pro](/models/deepseek/deepseek-v4-pro) | 163K | — | — | MIT | $0.00 / $0.00 | | [Qwen3-235B-A22B](/models/qwen/qwen3-235b-a22b) | 236K | 235B / 22B | — | Apache 2.0 | Free (open weights) | Motif 3 Beta fills a distinct niche: a high-parameter MoE model with a novel, fully proprietary architecture, free-to-use API access, and a long-context window — at the cost of a non-commercial license and beta-level stability. Compared to Apache-2.0 alternatives like Inkling or Qwen3-235B, it is more restricted commercially but provides comparable active-parameter efficiency with a unique architectural lineage. ## How Motif 3 Beta Performs on Real Agent Tasks An AAII score of 44 establishes Motif 3 Beta as a capable reasoning model in the mid-tier of the frontier — meaningfully above average but trailing the top proprietary models (Claude Fable 5, GPT-5.6 Sol, Kimi K3) that score in the 60–80+ range. The AAII incorporates agentic task categories including GDPval-AA v2 (knowledge work), τ³-Banking (financial SaaS workflows), and Terminal-Bench v2.1 (agentic coding/terminal use), which makes it a useful proxy for real agent deployments. What AAII does not capture is task-specific reliability across repeated runs. For builders evaluating Motif 3 Beta for production agent use, the priority is measuring how reliably it completes the specific domain workflows that matter — tool calling accuracy, long-horizon plan coherence, and error recovery — rather than treating the composite index as a go/no-go signal. Motif's 256K context and reasoning-model design make it a strong candidate for document-heavy agentic workflows; the beta label means regressions between checkpoints should be tracked before committing to a production dependency. ## Use Motif 3 Beta via Transformers ```python from transformers import AutoModelForCausalLM, AutoTokenizer import torch model_id = "Motif-Technologies/Motif-3-Beta" tokenizer = AutoTokenizer.from_pretrained(model_id, trust_remote_code=True) model = AutoModelForCausalLM.from_pretrained( model_id, trust_remote_code=True, torch_dtype=torch.bfloat16, device_map="auto", ) messages = [{"role": "user", "content": "Summarise the key clauses in this contract..."}] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, return_tensors="pt" ).to(model.device) outputs = model.generate(inputs, max_new_tokens=512) print(tokenizer.decode(outputs[0][inputs.shape[-1]:], skip_special_tokens=True)) ``` The model ships with custom modeling code — `trust_remote_code=True` is required. A dedicated vLLM serving guide is planned by Motif Technologies for the stable release. ## Related Models & Benchmarks - Compare: [Inkling](/models/thinkingmachines/inkling) · [Kimi K3](/models/moonshot/kimi-k3) · [DeepSeek V4 Pro](/models/deepseek/deepseek-v4-pro) · [Qwen3-235B-A22B](/models/qwen/qwen3-235b-a22b) - Benchmarks: [AA-Omniscience](/benchmarks/aa-omniscience) · [GPQA Diamond](/benchmarks/gpqa-diamond) · [Humanity's Last Exam](/benchmarks/humanitys-last-exam) · [Terminal Bench](/benchmarks/terminalbench) · [SciCode](/benchmarks/scicode) ## Frequently Asked Questions

What is Motif 3 Beta?

Motif 3 Beta is a 314B-parameter sparse Mixture-of-Experts reasoning model built from scratch by Motif Technologies, featuring a fully in-house architecture (GDLA attention, Grouped PolyNorm, modified mHC). It is a beta checkpoint of the Motif-3 series, released July 2026 ahead of the stable final version.

What is Motif 3 Beta's context window?

Motif 3 Beta has a 262,144-token (256K) context window — approximately 393 A4 pages in a single request. The long context is native to the model's base architecture.

How much does Motif 3 Beta cost?

During the beta period, Motif 3 Beta is available at $0.00 per 1M input and output tokens via Motif's model hub. Weights can also be downloaded from Hugging Face for free self-hosting. Commercial use requires a separate written agreement with Motif Technologies.

Is Motif 3 Beta open source?

The weights are publicly available on Hugging Face (no access request required), but the license permits personal, educational, and non-commercial research use only. Commercial use is prohibited without prior written permission from Motif Technologies, so it is not fully open source under standard OSI definitions.

How many parameters does Motif 3 Beta have?

Motif 3 Beta has approximately 314 billion total parameters, with approximately 13 billion active per token. The sparse MoE architecture activates 8 out of 384 routed experts (plus 1 shared expert) per forward pass.

Is Motif 3 Beta a reasoning model?

Yes. Motif 3 Beta uses extended thinking or chain-of-thought reasoning to work through complex problems before producing a final answer, as classified by Artificial Analysis.

What is Motif 3 Beta's knowledge cutoff?

Motif Technologies has not disclosed a specific training data cutoff date for the beta checkpoint.

--- *Specs and scores sourced from [Motif Technologies' Hugging Face model card](https://huggingface.co/Motif-Technologies/Motif-3-Beta) and [Artificial Analysis](https://artificialanalysis.ai/models/motif-0714). Last updated 2026-07-21.*

Motif 3 Beta — Training jobs

Training jobs