Together AI Integrations

Vertesia is a unified, low-code generative AI platform that enables enterprise teams to rapidly build, deploy, and operate GenAI applications and agents at scale. Designed for both business professionals and IT specialists, Vertesia offers a frictionless development experience, allowing users to go from prototype to production without extensive timelines or heavy infrastructure. It supports multiple generative AI models from leading inference providers, providing flexibility and preventing vendor lock-in. Vertesia's agentic retrieval-augmented generation (RAG) pipeline enhances generative AI accuracy and performance by automating and accelerating content preparation, including intelligent document processing and semantic chunking. With enterprise-grade security, SOC2 compliance, and support for leading cloud infrastructures like AWS, GCP, and Azure, Vertesia ensures secure and scalable deployments.

LFM2 is a next-generation series of on-device foundation models built to deliver the fastest generative-AI experience across a wide range of endpoints. It employs a new hybrid architecture that achieves up to 2x faster decode and prefill performance than comparable models, and up to 3x improvements in training efficiency compared to the previous generation. These models strike an optimal balance of quality, latency, and memory for deployment on embedded systems, allowing real-time, on-device AI across smartphones, laptops, vehicles, wearables, and other endpoints, enabling millisecond inference, device resilience, and full data sovereignty. Available in three dense checkpoints (0.35 B, 0.7 B, and 1.2 B parameters), LFM2 demonstrates benchmark performance that outperforms similarly sized models in tasks such as knowledge recall, mathematics, multilingual instruction-following, and conversational dialogue evaluations.

The leading open-source AI code assistant. You can connect any models and any context to create custom autocomplete and chat experiences inside the IDE Remain in flow while coding by removing the barriers that block productivity when building software. Accelerate development with a plug-and-play system that makes it easy to get started and integrates with your entire stack. Become a leader in AI by setting up your code assistant to evolve as new capabilities emerge. Continue autocompletes single lines or entire sections of code in any programming language as you type. Attach code or other context to ask questions about functions, files, the entire codebase, and more. Highlight code sections and press a keyboard shortcut to rewrite code from natural language.

gpt-oss-20b is a 20-billion-parameter, text-only reasoning model released under the Apache 2.0 license and governed by OpenAI’s gpt-oss usage policy, built to enable seamless integration into custom AI workflows via the Responses API without reliance on proprietary infrastructure. Trained for robust instruction following, it supports adjustable reasoning effort, full chain-of-thought outputs, and native tool use (including web search and Python execution), producing structured, explainable answers. Developers must implement their own deployment safeguards, such as input filtering, output monitoring, and usage policies, to match the system-level protections of hosted offerings and mitigate risks from malicious or unintended behaviors. Its open-weight design makes it ideal for on-premises or edge deployments where control, customization, and transparency are paramount.

gpt-oss-120b is a reasoning model engineered for deep, transparent thinking, delivering full chain-of-thought explanations, adjustable reasoning depth, and structured outputs, while natively invoking tools like web search and Python execution via the API. Built to slot seamlessly into self-hosted or edge deployments, it eliminates dependence on proprietary infrastructure. Although it includes default safety guardrails, its open-weight architecture allows fine-tuning that could override built-in controls, so implementers are responsible for adding input filtering, output monitoring, and governance measures to achieve enterprise-grade security. As a community–driven model card rather than a managed service spec, it emphasizes transparency, customization, and the need for downstream safety practices.

Wan 2.6 is Alibaba’s advanced multimodal video generation model designed to create high-quality, audio-synchronized videos from text or images. It supports video creation up to 15 seconds in length while maintaining strong narrative flow and visual consistency. The model delivers smooth, realistic motion with cinematic camera movement and pacing. Native audio-visual synchronization ensures dialogue, sound effects, and background music align perfectly with visuals. Wan 2.6 includes precise lip-sync technology for natural mouth movements. It supports multiple resolutions, including 480p, 720p, and 1080p. Wan 2.6 is well-suited for creating short-form video content across social media platforms.

Nemotron-3 Super is part of NVIDIA’s Nemotron 3 family of open models designed to enable advanced agentic AI systems that can reason, plan, and execute multi-step workflows across complex environments. The model introduces a hybrid Mamba-Transformer Mixture-of-Experts architecture that combines the efficiency of state-space Mamba layers with the contextual understanding of transformer attention, allowing it to process long sequences and complex reasoning tasks with high accuracy and throughput. This architecture activates only a subset of model parameters for each token, improving computational efficiency while maintaining strong reasoning capabilities and enabling scalable inference for large workloads. Nemotron-3 Super contains roughly 120 billion parameters with around 12 billion active during inference, accelerating multi-step reasoning and collaborative agent interactions across large contexts.

scribe is a self-hosted knowledge base written automatically by your tools. It reads Git history, Claude Code and Codex sessions, self-sent URLs, and drop files, then turns that work into a curated, cross-project wiki of plain Markdown stored in Git. Instead of making developers maintain a second brain or rebuild context whenever an agent session starts from zero, scribe captures decisions, fixes, evaluations, and the reasoning behind them so agents can query that memory before they act. Its pipeline runs on cron: it discovers projects, filters low-value noise with FTS5 before invoking an LLM, extracts grounded facts through bounded and two-pass workflows, and compiles them into entity-first pages with YAML frontmatter, wikilinks, backlinks, retrieval context, and typed relationships such as supersedes, contradicts, derived_from, specializes, and extends.

Adaption provides tools for building AI models that adapt as data and requirements change. Its web platform combines data preparation, synthetic-data generation, dataset exploration, and model training. Adaptive Data imports files and data from Hugging Face or Kaggle, with column mapping, deduplication, prompt transformation, augmentation, translation, and evaluation. Invent a Dataset creates instruction or preference data from a written objective. Visualize supports questions, interactive charts, and row-level inspection of text and tabular data. AutoScientist optimizes training data and model recipes against a goal, runs experiments, evaluates results, and exports adapted models. Workflows include fine-tuning, alignment, multilingual dataset expansion, and data-coverage analysis. Adaption also offers API access and a Python SDK for integrating these tools into AI-development workflows.