# Liquid AI releases open d1 models for multimodal edge decisions > Source: > Published: 2026-10-07 17:32:02+00:00 # Liquid AI releases open d1 models for multimodal edge decisions **Liquid AI released d1-3B for text and images and an experimental 600M model that pairs text with images or audio.** By [Ryan Merket](https://runtimewire.com/author/ryan-merket) ยท Published Primary source: [Hugging Face Newsroom](https://huggingface.co/blog/LiquidAI/open-d1) ## Why it matters The open-weight releases give developers a local alternative to API calls for structured tasks. Liquid AI's [d1-3B model card](https://huggingface.co/LiquidAI/d1-3b?ref=runtimewire) reports an average score of 74.1 across 11 public image benchmarks; the release gives no audio decision benchmark score. [Liquid AI](https://www.liquid.ai/?ref=runtimewire), co-founded by CEO [Ramin Hasani](https://www.liquid.ai/team/ramin-hasani?ref=runtimewire), released two open-weight models in its d1 decision family on October 7th. The [Hugging Face release](https://huggingface.co/blog/LiquidAI/open-d1?ref=runtimewire) introduces [d1-3B](https://huggingface.co/LiquidAI/d1-3b?ref=runtimewire), which accepts text and images, and experimental [d1-omni-600M](https://huggingface.co/LiquidAI/d1-omni-600M?ref=runtimewire), which accepts text paired with either images or audio. Developers can download and run the models themselves, or try demos in Liquid AI's [System One Arcade](https://huggingface.co/spaces/LiquidAI/system-one-arcade?ref=runtimewire). Hasani has pursued the idea of building models that work within the limits of real devices since Liquid AI's founding in 2023. Before co-founding [Liquid AI](https://www.liquid.ai/?ref=runtimewire), Hasani was a CSAIL postdoctoral associate and a principal AI and machine-learning scientist at Vanguard. His research includes decision-making in complex systems, relevant to models designed to return a choice or score instead of composing an answer in prose. Liquid AI introduced an API-hosted d1 model in a [company post on October 5th](https://www.liquid.ai/blog/d1-decision-model?ref=runtimewire) and said it planned to release open weights for upcoming models. Two days later, the Hugging Face post offered [d1-3B](https://runtimewire.com/models/huggingface/liquidai-d1-3b-fb3a1a604d63b0af) and [d1-omni-600M](https://runtimewire.com/models/huggingface/liquidai-d1-omni-600m-cdfd8291328b2175) as open-weight downloads. Liquid AI is making separate decision-model checkpoints available for local deployment alongside its API. ### A model built to return a decision Generative models produce tokens, even for narrow tasks such as sorting a support ticket or deciding whether an image shows a defect. Liquid AI's decision models instead process a state in a single forward pass and return structured answers. The release describes yes-or-no decisions, label selection and rubric-based scores as supported outputs. That format could suit routing, inspection and classification tasks where software needs an answer it can act on, rather than a paragraph it must parse. The two models use different foundations. d1-3B is built on Liquid AI's [LFM2.5-VL-3B](https://runtimewire.com/models/huggingface/liquidai-lfm2.5-vl-3b-7f26099f5ddfe449) vision-language model. d1-omni-600M uses the LFM2.5-Encoder-350M bidirectional encoder, with added vision and audio encoders. Liquid AI labels the smaller model experimental and says it remains under development; it did not report speed numbers for it. The new downloads let developers evaluate local deployment, while the October 5th release described an API-hosted model. Liquid AI's announcement says both new models are available on Hugging Face and includes code showing d1-3B loaded through Transformers. It also instructs users to enable `trust_remote_code=True`. The release calls the models open-weight but does not state a license. Both Hugging Face model cards list LFM 1.0; teams evaluating deployment should review those terms for their intended use. ### Company-reported benchmarks and latency In its [Hugging Face release](https://huggingface.co/blog/LiquidAI/open-d1?ref=runtimewire), Liquid AI reports a Decision Index 0.2.1 score of 48.57 for d1-3B, above the listed 4B and 9B models and Decider 35B-A3B at 47.11. For a separate comparison across seven public datasets, Liquid AI reports average scores of 82.9 for d1-3B and 78.4 for d1-omni-600M. Those figures exceed the listed Decider 4B and 2B averages, respectively. The October 7th announcement's benchmark tables do not include vision or audio scores. The [d1-3B model card](https://huggingface.co/LiquidAI/d1-3b?ref=runtimewire) separately reports an average score of 74.1 across 11 public image benchmarks, compared with 73.9 for its LFM2.5-VL-3B base model. Liquid AI says the Decision Index v0.3 vision split is private and that audio decision benchmarks remain an open problem. The latency figures are also company-reported. In the same [release](https://huggingface.co/blog/LiquidAI/open-d1?ref=runtimewire), Liquid AI reports that d1-3B answered one question in 16 milliseconds on a Jetson AGX Thor, 26 milliseconds on a Jetson AGX Orin and 50 milliseconds on a Jetson Orin Nano. On the Thor, the reported time for three questions was 20 milliseconds. These results are specific to the tested devices and workloads; they do not establish performance across other hardware or tasks. Liquid AI's d1 models extend its device strategy to structured decisions and let developers run models outside a hosted API. Whether their speed and quality hold up in production tasks remains to be tested, particularly for the experimental audio-capable model. Liquid AI said its $250 million Series A, announced in December 2024 and led by AMD, would support model development and edge and on-premise product readiness, according to its [funding announcement](https://www.liquid.ai/blog/we-raised-250m-to-scale-capable-and-efficient-general-purpose-ai?ref=runtimewire). The d1 release applies that stated goal to compact models that return machine-readable decisions locally. A separate [Hugging Face benchmark](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/blame/main/README.md?ref=runtimewire) tested d1 on September 30th across 400 cases and 2,000 decisions, reporting 0.742 accuracy; it did not directly replicate Liquid AI's seven-dataset comparison.