Engineering teams shipping with AI have a new bottleneck: validation. Code output has accelerated. Quality hasn't. Checksum closes the gap.
Checksum is a continuous quality platform with a suite of AI agents that handle testing end-to-end, at every stage of the development lifecycle. Where most tools wait for a human to trigger them, Checksum runs autonomously in the background, generating tests, executing them, and repairing failures without manual intervention. Seventy percent of test failures are resolved automatically through real-time auto-recovery.
The platform covers every layer: end-to-end UI flows via Playwright, API endpoint chains, and targeted CI tests scoped to exactly what changed in a PR. All tests land as real code in your repository and are delivered as standard Playwright, owned by your team.
Checksum is fine-tuned on 1.5+ million test runs and integrates natively with Cursor, Claude Code, and 100+ AI coding agents. Type /checksum and your coding agent's output gets tested before it ever reaches review. Generation and healing happen on Checksum's cloud infrastructure which means no LLM tokens consumed, no local resources required.
The result: test suites that stay green as the product evolves, fewer regressions reaching production, and release confidence that scales alongside AI output.
Learn more
Most AI video tools hand you a black box: closed weights, a subscription, and no way to see what is happening under the hood. LTX takes the opposite approach. Built by Lightricks, LTX is an open foundation model that generates and simulates across video, audio, and the physical world, and it puts the weights, the code, and the control in your hands.
At the center of the model is LTX-2.5, a 22B-parameter dual-stream diffusion transformer that produces native 4K video at up to 50 frames per second, with audio and video generated together in a single pass rather than stitched together afterward. Artificial Analysis, an independent benchmarking group, currently ranks LTX among the top three AI video models in the world.
You choose how you want to use it. Download the open weights and run LTX-2.5 on your own hardware. License the model for on-premise deployment backed by enterprise support. Or build directly on LTX Studio, the production suite that turns the model into a full creative workflow. Companies like ElevenLabs, Asteria Film Co., Magnopus, and NVIDIA already rely on LTX for their own work.
LTX is not built for one-off social clips. It is infrastructure for teams that generate motion, audio, and physical environments as part of their own products and pipelines.
Learn more
Aya
Aya represents a cutting-edge, open-source generative language model that boasts support for 101 languages, significantly surpassing the language capabilities of current open-source counterparts. By facilitating access to advanced language processing for a diverse array of languages and cultures that are often overlooked, Aya empowers researchers to explore the full potential of generative language models.
In addition to the Aya model, we are releasing the largest dataset for multilingual instruction fine-tuning ever created, which includes 513 million entries across 114 languages. This extensive dataset features unique annotations provided by native and fluent speakers worldwide, thereby enhancing the ability of AI to cater to a wide range of global communities that have historically had limited access to such technology. Furthermore, the initiative aims to bridge the gap in AI accessibility, ensuring that even the most underserved languages receive the attention they deserve in the digital landscape.
Learn more
Phi-2
We are excited to announce the launch of Phi-2, a language model featuring 2.7 billion parameters that excels in reasoning and language comprehension, achieving top-tier results compared to other base models with fewer than 13 billion parameters. In challenging benchmarks, Phi-2 competes with and often surpasses models that are up to 25 times its size, a feat made possible by advancements in model scaling and meticulous curation of training data.
Due to its efficient design, Phi-2 serves as an excellent resource for researchers interested in areas such as mechanistic interpretability, enhancing safety measures, or conducting fine-tuning experiments across a broad spectrum of tasks. To promote further exploration and innovation in language modeling, Phi-2 has been integrated into the Azure AI Studio model catalog, encouraging collaboration and development within the research community. Researchers can leverage this model to unlock new insights and push the boundaries of language technology.
Learn more
