Thomson Reuters releases Thomson-1.0-Small. A law and tax focused model
Summary
Thomson Reuters has released Thomson-1.0-Small, an AI model specialized in law and tax applications.
Similar Articles
thomsonreuters/Thomson-1.0-Small
Thomson-1.0-Small is an open-weight foundation model from Thomson Reuters, developed using continual learning to improve performance in legal, tax, and journalism domains with a focus on value sovereignty and efficiency.
Thomson Reuters Launches Its Own Frontier Model
Thomson Reuters has launched its first proprietary large language model, named Thomson, trained on a mix of open-source foundation and proprietary content to provide efficient, domain-specific AI for professionals.
@svpino: Tiny, specialized, and open models are the future! The TwiL-LM family of models is now available on HuggingFace for Tra…
TwiL-LM, a family of tiny specialized open models, is released on HuggingFace. The 3B version outperforms OpenAI's 120B gpt-oss on formal reasoning benchmarks and runs efficiently on consumer hardware.
Continual Learning of Frontier Models for SovereignAI. Tech Report + Open Weights Model [R]
The report introduces Thomson, a new general-purpose AI model trained through continual learning for sovereign AI, demonstrating competitive performance with efficient compute and addressing forgetting issues.
@natolambert: Thinky with a ~1T param, 41B active, apache-2 model Benchmarks are a clear step up from Nemotron Ultra (55B active), ne…
Thinky is a ~1T parameter mixture-of-experts model with 41B active parameters, released under Apache-2 license. It achieves new best results among American models, benchmark improvements over Nemotron Ultra, with omni-modal input.