Motif-3

1. Model Introduction

Motif 3 is a large-scale, decoder-only Mixture-of-Experts (MoE) language model with 314 billion total parameters and 13.2 billion parameters activated per token. It is built from the ground up by Motif Technologies following a fully in-house, proprietary design.

Motif 3 is built around Grouped Differential Latent Attention (GDLA), which integrates grouped differential attention with the compressed key–value representation of Multi-head Latent Attention. The architecture further incorporates modified manifold-constrained hyper-connections (mHC), Expert-Specific PolyNorm activations, and a Multi-Token Prediction (MTP) head to improve optimization stability, expert specialization, and inference efficiency.

The model is pretrained on approximately 12.5 trillion tokens spanning web documents, STEM, code, mathematics, multilingual content, and domain-specialized corpora, with additional emphasis on Korean, reasoning-intensive, legal, and financial data. Post-training combines general supervised fine-tuning, six RL-trained specialist teachers, a software-engineering teacher, and Multi-teacher On-Policy Distillation (MOPD) into a single unified model.

Key Features

• 🧠 Fine-grained sparse MoE — 384 routed experts with only 8 activated per token (plus 1 shared expert), providing a large expert pool at limited per-token compute.
• 📏 Native 256K context (262,144 tokens), trained with window-aware context parallelism.
• ⚙️ Novel architecture — GDLA attention, Expert-Specific PolyNorm, modified mHC, and a built-in MTP head enabling self-speculative decoding.
• 🌐 Multilingual & general-purpose, with a strong bytes-per-token tokenizer for English, Korean, code, and math.
• 🎯 Agentic strengths — particularly strong on long-horizon agentic tool use and terminal-based problem solving, with calibrated abstention on hallucination-sensitive evaluations.

2. Model Summary

3. Evaluation Results

For contextual comparison, Motif 3 is compared with strong open-weight models using scores reported on the corresponding benchmark leaderboards. All Motif 3 evaluations were performed with sampling temperature = 1.0, top-p = 0.95, and a maximum sequence length of 262,144 tokens. (*: public dataset only)

Motif 3 performs particularly well on agentic and tool-oriented benchmarks, while maintaining competitive performance across coding, mathematical reasoning, and general knowledge. On AA-Omniscience it pairs its accuracy with one of the highest non-hallucination scores, indicating a favorable balance between answering correctly and abstaining when unsupported.

4. Architecture

[!NOTE] The architecture and distributed training framework used for Motif 3 are available at MotifTechnologies/motif3-training-example.

Motif 3 is a fully in-house design and introduces several custom components (full details in the technical report):

• Grouped Differential Latent Attention (GDLA) — integrates grouped differential attention (asymmetric signal/noise heads with a token-dependent differential coefficient) with the compressed KV latent of Multi-head Latent Attention, plus a query-dependent output gate. Retains the expressive attention dynamics of differential attention while substantially reducing KV-cache requirements.
• Expert-Specific PolyNorm — replaces the SiLU gate with a learned polynomial normalization whose coefficients are learned independently per expert, reducing activation outliers while allowing each expert to specialize.
• Modified manifold-constrained hyper-connections (mHC) — replaces conventional residual additions with a doubly-stochastic (Birkhoff-polytope) mixing of 4 parallel residual streams; the post-mapping multiplier is annealed from 2 → 1 during pretraining to limit activation-outlier accumulation.
• Multi-Token Prediction (MTP) — a 1-layer MTP head (DeepSeek-V3 style) used as an auxiliary pretraining objective and enabling self-speculative decoding at inference.

5. Deployment — vLLM (Recommended)

[!Note]

• Tested on B200 and H200 GPUs.
• The model ships with a built-in MTP head (num_nextn_predict_layers=1), so it supports self-speculative decoding — add --speculative-config as shown below (num_speculative_tokens: 1 is optimal for this model).
• Supports online block-fp8 quantization with --quantization modelopt_blockfp8
• If you encounter any issues, please open an HF issue.

[!Tip] Looking for a smaller footprint? An NVFP4-quantized checkpoint is available at [Motif-Technologies/Motif-3-NVFP4](htt