ML Inference @ Waveforms AI
California
Joined September 2009
- Tweets47
- Following94
- Followers81
- Likes700
Hemant Jain retweeted
Excited to introduce Muse Realtime Avatar. ✨🤖
A state-of-the-art realtime Avatar model that brings your Muse to life at ~870ms response latency.
Coming soon to Muse and Muse Charm, where you can create your own character.
Blog: research.meta.ai/blog/bringi…
Hemant Jain retweeted
a meeting booked without touching a thing. no phone. no screen. just a conversation with your personal AI agent.
watch Muse do it live on glasses. #metaconnect
🤖 Made with AI
Hemant Jain retweeted
𝗪𝗲 𝗯𝗲𝗹𝗶𝗲𝘃𝗲 𝘁𝗵𝗲 𝗳𝘂𝘁𝘂𝗿𝗲 𝗼𝗳 𝗔𝗜 𝗶𝘀 𝗹𝗼𝗰𝗮𝗹, 𝗽𝗿𝗶𝘃𝗮𝘁𝗲, 𝗮𝗻𝗱 𝗽𝗲𝗿𝘀𝗼𝗻𝗮𝗹𝗶𝘇𝗲𝗱!
𝗖𝗼𝗯𝗼𝗹𝘁 — a free cross-platform AI assistant that runs entirely locally.
Try Cobolt: github.com/platinum-hill/cob…
#LocalAI #OpenSource #MCP #Ollama #Privacy
Hemant Jain retweeted
We’ve included detailed guides and a pre-built container..
Let accessible fine-tuning begin 🔥
GitHub repo: github.com/cohere-ai/cohere-…
Hemant Jain retweeted
@Cohere @CohereForAI has open-sourced a fine-tuning repo and container for easy customizing 🎉
This makes finetuning Command and Aya Expanse open-weight models on @huggingface far more accessible. 🔥
Hemant Jain retweeted
Command R fine-tuning just got easier with @cohere & W&B integration. Track experiments, visualize metrics, and optimize—all with one extra line of code. Check it out: wandb.ai/wandb_fc/product-an…
Hemant Jain retweeted
We’ve also developed a seamless integration with @wandb. This allows for enhanced monitoring and evaluation of fine-tuning experiments, providing greater visibility into training efficiency and outcomes.
docs.cohere.com/page/convfin…
Hemant Jain retweeted
We're excited to announce updated fine-tuning features alongside fine-tuning support for Command R 08-2024! These new features offer increased flexibility, control over data, and visibility into the fine-tuning process.
cohere.com/blog/fine-tuning-…
Hemant Jain retweeted
Hemant Jain, Engineering Manager in Finetuning at Cohere, will speak at the @awscloud Booth during @nvidiagtc on March 21 at 11:45 am PDT & 12:20 pm PDT. Come by Booth #708 to ‘Learn the art of model fine-tuning’ and ‘Why Embeddings are the best kept secret in GenAI’.
@IndiGo6E is a classic example of the lack of good engineers & no focus on user experience.
20 mins to get an OTP for a payment that never even asked for one and gets eventually rejected.
30 mins to get an email from their customer service for my flight details.
Do better!
Hemant Jain retweeted
We are excited to introduce our comprehensive fine-tuning suite. Businesses can now train and deploy a unique chat, reranking, or multi-label classification model in less than 30 minutes. txt.cohere.com/fine-tuning-s…
Hemant Jain retweeted
🚀 Discover how BlueDot turned complex API calls into easy natural language queries with our Classify & Rerank solutions. From 50% to 97% accuracy – a game-changer in global health intelligence. Read our story 📖 and see the power of LLMs in action! cohere.com/customer-stories/…
Hemant Jain retweeted
Congratulations @aidangomez on being chosen for the TIME AI 100!
Hemant Jain retweeted
Interested in learning about the challenges of serving fine-tuned large language model and ways to overcome them?
Join @hemanham this Thursday 1140am ET at the LLMs and the Generative AI Revolution virtual conference, organized by @AiInfra.
Register now,
community.ai-infrastructure.…
Hemant Jain retweeted
At Cohere, we implement the T-Few technique (Liu et al., 2022) for training and serving custom models.
In this blog post, we delve into the concept of T-Few finetuning, explore its benefits, and explain our implementation workflow.
txt.cohere.com/tfew-finetuni…
Hemant Jain retweeted
(1/15) This roundup highlights some interesting NLP papers and resources from 2022 highlighted by @forai_ml’s research community!
This post has been generated with Cohere. 🧵
txt.cohere.ai/best-nlp-resea…
Hemant Jain retweeted
Attending @PyData NYC? Join @CohereAI's @hemanham on Nov 10th, 11:45am EDT, where he'll give a "gentle intro" to real-world applications of large language models. No prior knowledge needed! Read more about the talk:
nyc2022.pydata.org/cfp/talk/…