🚀 DeepSeek-R1 now available on Together AI!. DeepSeek-R1 sets a new standard for open-source reasoning models, rivaling OpenAI-o1. It’s faster than closed-source alternatives, delivers the same top-tier performance across math, code, and reasoning tasks, and operates at a

🚀 DeepSeek-R1 distilled models are now available on Together AI:. ▶️ DeepSeek-R1 Llama 70B distilled – Surpasses GPT-4o with 94.5% on MATH-500 & matches o1-mini on coding (57.5% vs 53.8% LiveCodeBench). ▶️ DeepSeek-R1 Qwen 14B distilled – Outperforms GPT-4o in math (93.9% vs

Try DeepSeek-R1 distilled models today:. DeepSeek-R1 Llama 70B distilled:. DeepSeek-R1 Qwen 14B distilled:. DeepSeek-R1 Qwen 1.5B distilled:.

Read our blog post on all the different R1 models available on Together AI – all with full opt-out privacy controls and pay-per-token pricing:.

🔖 New Notebook: Serial Prompt Chain Agent!. In short, it involves:.1. Cleaning & extraction of PDF content LLM call.2. Outline generation LLM call.3. Script generation in JSON format.4. Text-to-Speech conversion. Check out how to build this simple but powerful chain!

📚 Together Cookbooks: Your guide to building with LLMs!. Step-by-step examples for:.▪️Agent workflows.▪️RAG systems.▪️LLM fine-tuning.▪️Search. Get started here!

RT @julien_c: which means it's already on the model page!!

RT @vipulved: @TheAIVeteran @togethercompute We provide strong privacy controls, eg: you can choose your prompt data is never persisted. Th….

🚀 New 100% free API endpoint for DeepSeek-R1 Llama 70B distilled. Start experimenting with the power of reasoning models today,. Available now on Together AI with full-opt out privacy controls. We host these models in our own data centers, and none of your data is ever sent

👉 Try DeepSeek-R1 Llama 70B distilled for free on Together AI:.

👉 Try it now on Together AI:

🚀Announcing Mistral Small 3 on Together AI!. This new 24B open-source model matches GPT-4o mini's performance and Llama 3.3 70B—at a fraction of the size. Perfect for:. 🤖 Chatbots & customer support.⚡ Fast processing of routine queries.🦾 Robotics & low-latency apps. Proud to

👉 Read more in our blog-post:

As an official launch partner, we’ve made Mistral Small 3 available on our high-performance serverless endpoint with:. ✅Fast inference speeds.✅OpenAI-compatible API.✅Pay-per-token pricing. Try it today:

RT @nutlope: There were 10k+ images generated from 2.5k users on loras​.dev in the last 24 hours!. Powered by the @togethercompute API – an….

🚀 DeepSeek-R1: Everything you need to know 🚀. Join @ZainHasan6 from Together AI next Thursday, Feb 6 at 9 AM PST for a deep dive into DeepSeek-R1. Learn about:.• How DeepSeek-R1 was built.• The training pipeline & distillation techniques.• Effective prompting for better

You can now @ the Together docs in @windsurf_ai directly to build powerful AI apps even faster!

Powered by Flux LoRAs through Together AI!.

Join us this Thursday in SF, for a conversation featuring @csris Together AI's founding VP of Engineering. Charles will dive into the world of compute and share insights on scaling AI infrastructure, balancing hardware advancements with algorithmic breakthroughs, and the future.