Results for “agents”

…when compilation fails. This is particularly critical for agents, according to @gabriel_syme. Link to '100x Speedup Full Stack Transformer Inference Optimization': @atgctg posted a link to an article titled 'Towards 100x Speedup: Full-Stack…

…LlamaIndex, and Pinecone to the emerging field of autonomous agents like Auto-GPT and BabyAGI (must-read recap from Lilian Weng here) Research/Progress: On top of this, the sheer volume of papers and models and techniques published each day…

…Types and Uses: @problem9069 asked about different types of agents, such as OpenAITools and OpenAIFunctions, elaborating on intended model types and features. They questioned whether learning about all the types is necessary or if there's a…

…sensitive models are discussed, in under 12 hours using AI agents (@lizzadwoskin). Sarah Guo called the timing "savage" (@saranormous). Fable 5.5 expectations. It has reportedly rolled out to enterprise (@kimmonismus), and commentators…

…highlighted by users. A clarification from @solbus on how agents store and utilize uploaded files as 'knowledge' was noteworthy. "Uploaded files were stored as 'knowledge' for the agent's reference but did not continually modify their base …

…model for metabolic prediction tasks: @GoogleResearch. Agents, safety, and the OpenAI/Hugging Face incident OpenAI published its technical report on the Hugging Face incident: @OpenAI. The company also highlighted that the behavior came…

…discussed in broader circles and noted the integration of AI agents with music generation, while @pradeep1148 shared a YouTube video demonstrating how to fine-tune the Gemma model using Unsloth. Links mentioned: Oppenheimer Oppenheimer…

…framework. Folks started realizing that some of the cases where agents were failing were due to bad problem setups rather than just to models being dumb. So folks at OpenAI did a pretty extensive human data campaign hiring like almost a…

…and @solbus. The issue seemed to persist across custom agents for @stevenramenby, prompting further help requests in the specified channels. ChatGPT Down Experience: Users @zappymonster, @7877, @andrew.lol, @gardener_windson, @acaba1806,…

…collaboration Harrison from LangChain is promoting “Ambient” agents - always-on, passive, proactive, with cron/stateful orchestration/human-in-the-loop “interrupts” All different in nuanced ways, but A-SWE agents are now likely to be the…

…assessing the factuality of long-form LLM responses using LLM agents as automated evaluators through Search-Augmented Factuality Evaluator (SAFE). Stepwise Direct Preference Optimization (sDPO): @_akhaliq shared a paper proposing sDPO, an…

…launched Not Diamond Code, a router for long-horizon coding agents that selects both model and reasoning effort per step, claiming 20–65% cost reduction without quality loss. Similar themes showed up in @cognition, where Devin Fusion became…