We are excited to announce the public beta for Glaive. Starting today, developers and product builders can sign up and use Glaive to build and improve use case specific models by generating high quality synthetic datasets. Try it out here:

huggingface link to dataset:

Introducing Grok Imagine.

Today, we are releasing a synthetic dataset containing 22M+ reasoning traces for general purpose prompts across various domains. We noticed a lack of large datasets containing reasoning traces for diverse non code/math topics like social and natural sciences, creative writing,

RT @NousResearch: Introducing a new open dataset release, Hermes Function Calling V1, the datamix that gave Hermes 2 Pro its tool use and s….

RT @cognitivecompai: @TheEricHartford and Cognitive Computations present dolphin-2.9.4-llama3.1-8b. Now with extra SystemChatMultilingual,….

You can now finetune the new Llama 3.1 models by @AIatMeta on custom synthetic datasets on Glaive. Performance on our internal finetuning benchmarks looks extremely promising, go ahead and try it out.

RT @naklecha: we worked with groq to train the sota open source function calling model, yes literally the best! + we rank #1 on the berkele….

RT @mattshumer_: Groq function calling is incredible. Powered entirely by @GlaiveAI synthetic data!. Instant tool-use is going to be massiv….

RT @RickLamers: I’ve been leading a secret project for months … and the word is finally out!. 🛠️ I'm proud to announce the Llama 3 Groq Too….

RT @GroqInc: Today we launched the Llama-3 Tool Use 8B and 70B models! It’s fast and capable at #1 on the BFCL ahead of all other open-sour….

With this announcement, we are also releasing Llama-3-8B-RAG-v1, a model trained for RAG use cases built using the Glaive platform.

RT @csahil28: Really excited to share that @glaiveAI has raised a $3.5M seed round led by @sparkcapital with participation from @villageglo….

RT @csahil28: Excited to announce that I'm building @GlaiveAI , helping companies train use-case specific small language models with the he….