Model library / Motif-Technologies
Motif 3 is a 314.8 billion-parameter decoder-only mixture-of-experts language model with 262 k token context, released under MIT.
Mixture of experts, count not published
Context length in tokens, as published
Weights at the released precision
Licence declared on the repository
Overview
Motif 3 is a decoder-only mixture-of-experts language model containing 314.8 billion total parameters, of which 13.2 billion are activated per token. Motif Technologies describes it as built around Grouped Differential Latent Attention and specialised components to improve optimisation stability and expert specialisation. The model targets general-purpose multilingual tasks, with particular emphasis on long-horizon agentic tool use, coding, mathematics and legal or financial reasoning.
Deploying Motif 3 requires handling a 314.8 billion-parameter model stored in bfloat16 format. It supports a native context window of 262 144 tokens and is released under the MIT licence, allowing unrestricted use. The model uses a sparse MoE architecture with an unspecified number of experts, and is compatible with vLLM on B200 or H200 GPUs, offering self-speculative decoding and block-fp8 quantisation.
Specification
| Repository | Motif-Technologies/Motif-3 |
| Publisher | Motif-Technologies |
| Published | 7 August 2026 |
| Architecture | MotifForCausalLM |
| Model type | Motif |
| Parameters | 314,841,775,750 |
| Layers | 53 |
| Hidden size | 4,096 |
| Attention heads | 80, 16 key/value heads |
| Context length | 262,144 tokens |
| Vocabulary | 220,160 tokens |
| Weight format | bfloat16 |
| Licence | mit |
| Base model | Motif-Technologies/Motif-3-Base |
Memory
Parameter count multiplied by the bytes each format uses. The key-value cache comes on top.
| Precision | Weights | Notes |
|---|---|---|
| bf16 | 586.4 GiB | Full precision as released |
| fp8 | 293.2 GiB | 8-bit, near-lossless on most models |
| int4 | 146.6 GiB | 4-bit, smallest footprint |
The cache costs 636 KiB per token at 16-bit, so the full 262,144-token context of one request needs about 159.0 GiB. Concurrency multiplies that number, not the weights.
That figure is an upper bound: this model uses compressed key-value attention, so the real cache is a fraction of the figure above.
Hardware
At the released precision, with 90% of the card given to the server and 2.5 GiB kept for the runtime. The context column is what is left for the cache on that setup.
| GPU | Cards needed | Free for cache | Context it holds |
|---|---|---|---|
| RTX 4090 24GB | more than 8 | — | — |
| RTX 5090 32GB | more than 8 | — | — |
| L40S 48GB | more than 8 | — | — |
| A100 80GB | more than 8 | — | — |
| H100 80GB | more than 8 | — | — |
| RTX PRO 6000 96GB | 8 | 84.8 GiB | about 139,746 tokens |
| H200 141GB | 8 | 408.8 GiB | the full 262,144 tokens |
Serving
A vLLM launch line for the shape above. Check the model card for a runtime the publisher recommends.
vllm serve Motif-Technologies/Motif-3 \ --tensor-parallel-size 8 \ --max-model-len 139264 \ --gpu-memory-utilization 0.90
Questions
The weights take about 586 GiB at the released precision. No single card in the table above holds that, so it needs several GPUs or a lower precision.
The repository declares mit. Read the licence text before commercial use: the name alone does not tell you what is allowed.
The configuration allows 262,144 tokens. Whether the whole window is usable depends on the memory left for the cache, which the hardware table works out per card.
This page is a reference, and Aiqre does not serve this model today. We deploy open-weight models on dedicated EU hardware on request, so if you want an OpenAI-compatible endpoint for it, email [email protected] with the model name and roughly what volume you expect.
Also in the library
Nex-N2.5-Pro is a 396.8 billion-parameter MoE language model designed for long-horizon, agentic tasks with multimodal and compute…
GigaChat 3.5 Reasoning is a 438 billion-parameter mixture-of-experts language model trained with online RL for improved reasoning…
K2-Horizon-375B-A23B is an open-weight sparse MiE language model from IFM with 379 B parameters, 23 B active per token and a 512 …
DeepSeek-V4-Flash-0731 is a 304.2 billion-parameter, mixture-of-experts language model with speculative decoding for agentic task…
Hosting
Tell us the model and roughly what volume you expect. We reply with a price and an OpenAI-compatible endpoint, with your prompts never stored.