Laya
PulseAugur coverage of Laya — every cluster mentioning Laya across labs, papers, and developer communities, ranked by signal.
- instance of System One 90%
- used by Convai Innovations 90%
- developed by Convai Innovations 90%
- instance of Convai Innovations 90%
- developed TypeSafe.ai 90%
- instance of Jevíčko 70%
- developed by Lightbend Inc. 70%
- competes with Lightbend Inc. 70%
- developed by System One 70%
- uses Lightbend Inc. 70%
- instance of TypeSafe.ai 70%
- competes with System One 70%
19 day(s) with sentiment data
-
New Decision Model Enables Direct LLM Weight Editing Without Training
A new method allows for direct editing of large language model weights without traditional training, using a decision model inspired by MIT research. This technique, demonstrated with the Qwen 3.5 0.8B model, distills k…
-
Research reveals order sensitivity in Laya decision model
A new research paper titled "Maya a1: Do the Options Read the Model, or Does Their Position? Order Sensitivity in a Non-Autoregressive Decision Model" investigates the order sensitivity of the Laya model, a non-autoregr…
-
AI model performance and infrastructure costs discussed across multiple posts
A series of posts on Mastodon discuss the infrastructure and performance of AI models. One post details the cost of an MCP search service, estimating it at $7 per month for personal use and $8,400 for a two-region Azure…
-
New protocol reveals demographic bias in System One decision models
A new protocol, "Swap, Replicate, Read the Mass," has been developed to audit demographic bias in System One decision models. The protocol involves running models on benchmarks like BBQ, CrowS-Pairs, and SilverSpoon, wi…
-
New benchmark S1CogBench evaluates System One decision models on cognitive tasks
A new benchmark, S1CogBench, has been developed to evaluate the decision-making capabilities of System One decision models, such as Jev and Laya, on cognitive tasks. Unlike traditional language models, these systems pro…
-
Developer Rebuilds Karpathy's HackerTube as OpenTube, Measures Performance
A developer has recreated Andrej Karpathy's HackerTube project, naming their version OpenTube, to filter YouTube links from Hacker News comments. The project uses a two-stage model cascade, starting with a smaller model…
-
Local classifiers benchmarked against NLI cross-encoder; multi-agent LLM limitations explored
A user benchmarked a committee of three local classifiers (Laya, VegaML, and Arc Decide) against a DeBERTa-v3-small NLI cross-encoder on their CPU. The experiment aimed to determine when voting mechanisms improve classi…
-
AI agent apps run entirely in-browser using Laya decision model
A developer has created four AI agent applications that run entirely within a web browser using Laya, a 421M-parameter decision model from ConvAI Innovations, and WebGPU. These applications require no server or API keys…
-
LOKĀNTRA platform uses multi-agent AI to simulate societal reactions
LOKĀNTRA is a new AI-powered platform designed for multi-agent simulation and swarm intelligence. It transforms real-world documents into interactive knowledge graphs, enabling users to simulate how different AI agents …
-
AI models show mixed performance on rheumatology diagnostic tasks
A new study evaluated the performance of several AI models, including Decision and System One models, on rheumatology examination questions and diagnostic case vignettes. The models were tested on two benchmarks: the Rh…
-
New Jev-style model offers low-latency decisions for satellite operations
A new research paper introduces SatOps-Decision, a method designed for low-latency decision-making in satellite communication operations. This approach utilizes a Jev-style general-purpose discriminative decision model,…
-
Alaska Senate race faces 'same-name' controversy, drawing Trump's ire · 1 source tracked
A political controversy has emerged in Alaska's Senate race involving two candidates named Dan Sullivan, drawing attention from former President Trump. The situation is complicated by allegations that Democratic operati…
-
Laya AI model tested for insurance workflow decisions
The article discusses the application of Laya, an AI model, to insurance workflow decisions, focusing on its performance in terms of speed, token usage, cost, and accuracy. It highlights that not all AI-driven decisions…
-
DeepOpen framework achieves high accuracy on banking and intent classification tasks
Researchers have introduced DeepOpen, a framework for decision-oriented intent classification tailored for specific datasets like BANKING77 and CLINC150. The system achieved a high accuracy of 94.12% on BANKING77 using …
-
ELMDE system slashes LLM decision latency to milliseconds
A new system called the Equilibrium Latent-MoE Decision Engine (ELMDE) has been developed to significantly speed up decision-making processes in language models. Unlike traditional token-by-token decoding, ELMDE process…
-
Local AI model fine-tuning shows inconsistent replication
A fine-tuned local AI model, Laya, was tested against a hosted model (Jev) on a client's pipeline output. While Laya achieved comparable precision, it struggled with specific rule-based matches. Subsequent retraining at…
-
Agentic AI uses probabilistic decision models like Jev and Laya
The article delves into agentic AI, focusing on decision models that respond with probabilities rather than text. It explores the inner workings of such systems, using 'Jev' as an example of a model that answers typed q…
-
Jevíčko vs. Laya: Analyzing the true cost of 'free' AI models
The article compares two AI models, Jevíčko and Laya, to determine which offers better value despite initial appearances of being free. It suggests that when considering factors beyond the initial download, such as host…
-
AI decision APIs emerge, value shifts to structure over models
Several companies have recently launched APIs for fast, structured decision-making, including TypeSafe, OpenAI, Cloudflare, Perplexity, and Amazon. These offerings aim to provide quicker and cheaper alternatives to trad…
-
Jev, Drex, and Laya: System-One Decision Models Compared for Medical and General Use
A comparative study published on arXiv highlights the reliability of Jev, a system-one decision model, in medical contexts, finding it rivals leading models on text-driven tasks but struggles with numerical ones. The re…