Motif Technologies · Available now
Motif 3 price per reply (2026): Motif Technologies’ open-weights model
Motif 3 is Motif Technologies’ open-weights Mixture-of-Experts language model with 314B parameters and a 262,144-token context window, listed on OpenRouter on November 3, 2026 at $0.30 per million input and $1.20 per million output tokens, while Motif has published no API price of its own.
Use it now
Chat
≈ 1 credit per reply · See all models
New accounts get 5 free credits once, valid for 7 days. They work on the lower-priced models here; the priciest ones need credits you bought.
Start chatting — 5 free credits- Maker
- Motif Technologies
- Type
- Text
- Released
- No date given
- OpenRouter id
- motif-technologies/motif-3
Per million tokens
Price
| - | Input | Output | Source |
|---|---|---|---|
| Via OpenRouter (motif-technologies/motif-3) | $0.30 | $1.2 | openrouter.ai |
Checked
Facts, with sources
- The Hugging Face model card links "Motif 3: Technical Report" (arXiv 2608.09119), which arXiv shows as submitted on August 10, 2026. We found no page by Motif Technologies that states the model’s release date, so this page gives none.Source: arxiv.org · checked 2026-11-17 · second-hand
- Motif’s model card calls Motif 3 a decoder-only Mixture-of-Experts language model with about 314B total and 13.2B active parameters, "built from the ground up by Motif Technologies". It was pretrained on about 12.5T tokens, with extra emphasis on Korean, reasoning, legal and financial data; the card tags English and Korean and says it is multilingual and general-purpose.Source: huggingface.co · checked 2026-11-17
- Model card specs: native context length 262,144 tokens (256K), 53 layers (2 dense, 51 mixture-of-experts), 384 routed experts with 8 active per token plus 1 shared expert, vocabulary of 220,160 tokens, bfloat16 weights, and a 1-layer multi-token-prediction head that allows self-speculative decoding. The card lists text generation only.Source: huggingface.co · checked 2026-11-17
- Motif’s own figures, not tested by us: the card lists SWE-Bench Verified 76.2, Terminal-Bench 2.1 74.9, GPQA Diamond 83.4 and HLE 37.0. It says the model is "particularly strong" on long-horizon agentic tool use and terminal-based problem solving. The scores for the other models in its table are copied by Motif from public leaderboards.Source: huggingface.co · checked 2026-11-17
- The card says the model is released under the MIT License and that anyone can download the weights without an access request; Motif asks, but does not require, a mention such as "Built with Motif 3". We did not open the LICENSE file or the weight files ourselves.Source: huggingface.co · checked 2026-11-17
- Motif’s homepage offers "Start AI Chat" (chat.motiftech.io, with log-in and sign-up) with Motif 3 in the chat box; its "Get API Key" link is marked "Coming soon" and the homepage lists no prices. We found no Motif price list, so the maker price on this page is left empty.Source: motiftech.io · checked 2026-11-17
- For self-hosting, the model card recommends vLLM (Motif publishes a vLLM fork and a Docker image, ghcr.io/motiftechnologies/vllm:v0.26.0-motif3), tested on H200 and B200 GPUs, and points to a smaller NVFP4-quantized checkpoint, Motif-3-NVFP4. The card lists Hugging Face repo Motif-Technologies/Motif-3, with a separate Motif-3-Base model it was built from.Source: huggingface.co · checked 2026-11-17
- OpenRouter lists motif-technologies/motif-3 at $0.30 per million input tokens, $1.20 per million output tokens and $0.06 cache read, with a 262,144-token context window, up to 235,929 output tokens, text input and text output only, and support for tools, structured outputs and a reasoning parameter that is not mandatory (checked November 17, 2026).Source: openrouter.ai · checked 2026-11-17 · second-hand
- OpenRouter’s endpoint list for the model shows one host, KT, with quantization "unknown" and the same $0.30 / $1.20 price (checked November 17, 2026). OpenRouter’s list shows the model as created on November 3, 2026 Korea time (21:35 UTC on November 2); this is a listing date, not a release date from Motif.Source: openrouter.ai · checked 2026-11-17 · second-hand
Nearest available
Models you can use now instead
DeepSeek V4.1 Flash
DeepSeek$0.30 / $1.2 (per million tokens)
- DeepSeek released V4.1-Flash on September 10, 2026.Source: api-docs.deepseek.com · checked 2026-11-11
- Only the Flash version of V4.1 exists: DeepSeek has published no V4.1 Pro, and the Pro model on its price page is still deepseek-v4-pro (V4-Pro-0813).Source: api-docs.deepseek.com · checked 2026-11-11
- DeepSeek’s pricing page says the API model deepseek-flash is now served by the V4.1-Flash-1017 snapshot.Source: api-docs.deepseek.com · checked 2026-11-11
GLM 5.3
Z.ai$0.08 / $3.6 (per million tokens)
- A secondary report (DataNorth) says Z.ai released GLM-5.3 on August 14, 2026; Z.ai’s docs page calls it "Z.ai’s latest flagship model" but states no date.Source: datanorth.ai · checked 2026-11-15 · second-hand
- Z.ai’s docs say GLM-5.3 advances complex software engineering and agent capabilities, with an emphasis on coding and cybersecurity performance.Source: docs.z.ai · checked 2026-11-15
- Context window 1M tokens, maximum output 128K tokens; text-only input.Source: docs.z.ai · checked 2026-11-15
FAQ
Questions
What is Motif 3?
Motif 3 is a language model from Motif Technologies with about 314B total and 13.2B active parameters, built as a Mixture-of-Experts design. Motif’s model card calls it multilingual and general-purpose, with extra Korean, reasoning, legal and financial data in pretraining. You can chat with it on this page on credits, without a Motif account.
How much does Motif 3 cost?
Motif has not published an API price: its homepage shows "Get API Key" as coming soon and lists no prices. OpenRouter lists motif-technologies/motif-3 at $0.30 per million input tokens and $1.20 per million output tokens (cache read $0.06), from one host, KT (checked November 17, 2026). On this site a typical reply to this model costs about 1 credit (a credit is about $0.01 in the $5 pack), billed by its real token cost, at least 1 credit.
How do I use Motif 3?
Three ways: the Motif chat at chat.motiftech.io (log-in or sign-up needed), OpenRouter with the id motif-technologies/motif-3, or this page’s chat box on credits. Motif’s own API is marked "Coming soon". You can also download the weights from Hugging Face (Motif-Technologies/Motif-3) and self-host them; Motif recommends vLLM.
What is the context window of Motif 3?
Motif’s model card gives a native context length of 262,144 tokens (256K). OpenRouter lists the same 262,144 tokens, with up to 235,929 output tokens, and text input and output only.
Is Motif 3 open source?
Motif’s model card says the model is released under the MIT License and that anyone can download the weights without asking for access. We did not open the licence file or the weights ourselves, so read the licence before commercial use.
Is Motif 3 good at coding and agents?
Motif says it is particularly strong on long-horizon agentic tool use and terminal-based problem solving. Its own figures, not tested by us, include 76.2 on SWE-Bench Verified and 74.9 on Terminal-Bench 2.1. Try your own prompts in the chat box to see how it handles your task.
Is this the official Motif website?
No. This is an unofficial page, not affiliated with or endorsed by Motif Technologies. Facts link to Motif’s own pages where possible; OpenRouter and other outside sources are marked as such.