Your data
Define approved inputs, source access, and handling requirements.
- Isolation by design
- No training on your data
- Data residency options
Build for Greek and multilingual enterprise applications with full control over your data and deployment.
Before starting the pilot, agree the workflow owner, approved sources, and evaluation criteria.
What needs to happen before our pilot begins?
English and Greek speech-to-text, including mixed Greek–English speech, for meetings, calls, and recordings.
API access available on request.
ASR K1 preview · Latest results · Default datasets · Verified View leaderboard
Greek-first multimodal chat with English support and formal-to-informal register control. Choose a release for your runtime and modality.
A Greek and English language model focused on Modern Greek morphology, including correctly inflected word forms.
A Greek-language model with publicly available weights and published Greek-language evaluation results.
Greek and English reasoning with retained vision capabilities. Available as the original fine-tune and a reinforcement-learning refinement.
A Greek and English reasoning model with separate reasoning and answer channels, trained to reason in the language of the question.
Greek and English reasoning built on a hybrid mixture-of-experts architecture, with reasoning matched to the question’s language.
Base-model terms also apply.
Greek and English reasoning with language-matched responses and a multi-token prediction head for speculative decoding.
Base-model terms also apply.
Dense Greek and English embeddings for semantic search and retrieval-augmented generation. Represent queries and documents for relevance matching.
A Greek and English cross-encoder that reranks retrieved passages by relevance to the query, refining the context passed to a language model.
Llama 3.2 Community License also applies; see the model card and base-model terms.
Document-grounded answer generation in Greek and English, trained to cite sources and abstain when the supplied evidence is insufficient.
See the model card and base-model terms.
Greek and English speech recognition, plus Greek-to-English speech translation. Standard and MLX weight releases are available.
LIBERO Greek v1: an experimental Greek and English vision-language-action policy for robot manipulation, trained and evaluated in simulation.
See the model card and base-model terms.
Public weights are available under the terms of each release. Model cards describe supported formats, evaluation methods, limitations, and applicable base-model licenses.
Support enterprise questions with grounded responses and clear paths to human review.
Explore in a pilotWhat needs to happen before our pilot begins?
Agree the owner, approved sources, and evaluation criteria.
A shared checklist for the discovery team.
Evaluate against representative tasks before choosing a model.
Review failure cases, refine the setup, and repeat with the same evaluation criteria.
Define approved inputs, source access, and handling requirements.
Scope roles, tools, and review boundaries for the application.
Align the model runtime with operational requirements.
Identify the evidence needed to inspect quality and behavior.
// Illustrative integration — contract to be confirmed
const answer = await sophea.knowledge.ask({
question: "What are the next steps?",
sources: approvedSources,
identity: currentUser,
includeCitations: true
});
// Review evidence before taking action
renderAnswer(answer.text, answer.citations);"text": "Confirm the workflow owner and approved sources."
"citations": ["Pilot brief · p. 2"]