Together · Blogs

Blogs listings from Together on TechiSeek. 208 public listings.

• Announcing native availability of NVIDIA Nemotron 3 Nano, NVIDIA’s latest reasoning model (Together)
  - NVIDIA Nemotron 3 Nano, the company’s newest reasoning model, is now available on Together AI, the AI Native Cloud — combining big-model…
• Together AI and Pearl Research Labs Team Up to Reduce the Cost of AI Inference (Together)
  Together AI’s mission is to make AI accessible and widely distributed. Today, we are excited to announce an exclusive partnership with Pear…
• New in Together GPU Clusters: Autoscaling, observability, and self-healing (Together)
  AI infrastructure has quietly become production infrastructure. Teams are no longer experimenting with a handful of GPUs.
• Together AI Powers Pioneers at GTC: NVIDIA Blackwell GPUs, Instant GPU Clusters, and A Full-Stack (Together)
  This week, as an NVIDIA GTC Gold Sponsor, we're taking the wraps off our latest advancements to Together GPU Clusters.
• Together AI announces strategic partnership with Moonshot AI to natively serve Kimi models (Together)
  Under this partnership, Together AI becomes a launch platform for Moonshot's model releases, starting with Kimi K3 and extending to every o…
• Consistency diffusion language models: Up to 14x faster inference without sacrificing quality (Together)
  We introduce consistency diffusion language models (CDLM), which accelerates diffusion language model inference by combining consistency-ba…
• Minions: embracing small LMs, shifting compute on-device, and cutting cloud costs in the process (Together)
  We present Minions, a method that shifts a substantial portion of LLM workloads to consumer devices by having small on-device models collab…
• Together Code Interpreter: execute LLM-generated code seamlessly with a simple API call (Together)
  Today, we introduced Together Code Interpreter (TCI): an API that lets you run code generated by LLMs and get an instant response.
• NeurIPS 2022: Overcoming communication bottlenecks for decentralized training (2/2) (Together)
  In our previous blog post, we introduced our work on scheduling foundation model training over heterogeneous networks.
• From Zero to One: Building An Autonomous and Open Data Scientist Agent from Scratch (Together)
  While straightforward to implement, our agent performs well on several different use cases and benchmarks.
• DeepSWE: Training a Fully Open-sourced, State-of-the-Art Coding Agent by Scaling RL (Together)
  Through a joint collaboration between the Agentica team and Together AI, we introduce DeepSWE-Preview, a reasoning-enabled coding agent tra…
• Announcing v1 of our Python SDK (Together)
  The Together AI Python SDK is officially out of beta with the v1 release! It provides great OpenAI compatible APIs to: - Run inference on c…
• DeepSeek-V4 Pro now available on Together AI (Together)
  - Deep Seek V4 Pro on Together AI: Deep Seek V4 Pro is now available on Together AI with a 512 K-token context window for long-context reas…
• AWS Marketplace now offering Together AI to accelerate enterprise AI development (Together)
  Today, we're excited to announce that Together AI is now available on Amazon Web Services (AWS) Marketplace, and has joined the AWS Partner…
• Introducing Together AI Chief Scientist Tri Dao, as he releases FlashAttention-2 to speed up model (Together)
  We are thrilled to announce our new Chief Scientist, Tri Dao. Tri works at the intersection of machine learning and systems, and his work h…
• Capacity without conflict: A guide to multi-tenant GPU cluster design for AI-native teams (Together)
  Multi-tenant GPU clusters let AI-native companies share compute capacity across teams without sacrificing isolation or control.
• Supercharging NVIDIA H200 and H100 GPU Cluster Performance With Together Kernel Collection (Together)
  Today, we're excited to announce that the NVIDIA H200 Tensor Core GPU will soon be available in our Together GPU Clusters.
• ParallelKernelBench: Frontier LLMs can't write fast multi-GPU kernels (yet) (Together)
  LLMs have gotten surprisingly good at writing GPU kernels[1][2][3], but almost all current benchmarks measuring that progress are single-GP…
• On-demand dedicated endpoints: run inference with unmatched price-performance & control at scale (Together)
  Today, we are excited to announce on-demand Dedicated Endpoints—now available with up to 43% lower pricing, delivering the best price-perfo…
• Together Evaluations now supports comparing top commercial APIs vs. open source models (Together)
  Together Evaluations now supports OpenAI, Anthropic, and Google models for comprehensive benchmarking.
• DeepCoder: A Fully Open-Source 14B Coder at O3-mini Level (Together)
  Through a joint collaboration between the Agentica team and Together AI, we release Deep Coder-14 B-Preview, a code reasoning model finetun…
• Introducing Dedicated Container Inference: Delivering 2.6x faster inference for custom AI models (Together)
  Dedicated Container Inference lets teams deploy custom generative media models — like video generation, avatar synthesis, and image process…
• What is an AI Native Cloud? (Together)
  Over the last few years of powering and partnering with the fastest scaling AI-native companies, we have come to realize they need a differ…
• Monarch Mixer: A new model architecture for increased efficiency (Together)
  Transformers are the workhorse architecture behind modern foundation models. Our team at Together AI has released optimizations like Flash…
• Plan, divide, and conquer: How weak models excel at long context tasks (Together)
  The Intuition: Don't ask one genius to read a library in an hour. Our research, "When Does Divide and Conquer Work for Long Context LLM?" (…
• Bringing 100,000 GPUs to Europe (Together)
  This marks our largest infrastructure deployment in Europe. Initial rollouts begin in late 2025 with large-scale buildouts planned through…
• Introducing Together Chat: use DeepSeek R1 for free, hosted in North America (Together)
  Here's a list of some of the models that we support: - Deep Seek R1 with web access for difficult questions - Llama 3.3 70 B / Qwen 2.5 72…
• Together AI Awarded ClusterMAX™ Gold Rating by SemiAnalysis (Together)
  Semi Analysis has introduced the ClusterMAX™ Rating System for GPU clouds – and we are honored that our Together GPU Clusters have been rat…
• Deploy and inference any model from HuggingFace (Together)
  Something real is shifting in how developers work. Agents open up work that used to be off-limits, not because it was technically impossibl…
• Together MoA (Together)
  We introduce Mixture of Agents (MoA), an approach to harness the collective strengths of multiple LLMs to improve state-of-the-art quality.
• Together AI becomes NVIDIA Cloud Partner to bolster accelerated AI offerings (Together)
  Expanding our collaboration with NVIDIA, we are proud to announce Together AI is now an NVIDIA Cloud Partner in the NVIDIA Partner Network.
• Our $102.5M Series A (Together)
  At Together AI, we believe the future of AI is open source. So we are creating a comprehensive cloud platform to allow developers everywher…
• Building your own RAG application using Together AI and Langchain (Together)
  Together AI provides the fastest cloud platform for building and running generative AI. As part of a series of blog posts about the Togethe…
• Fine-tuning API: Introducing long-context training, conversation data support and more configuration (Together)
  As organizations race to gain competitive advantages with generative AI, fine-tuning Large Language Models has become critical for enhancin…
• FlexGen: High-throughput generative inference of large language models with a single GPU (Together)
  The high computational and memory requirements of large language model (LLM) inference make it feasible only with multiple high-end acceler…
• Together AI Launches Speech-to-Text: High-Performance Whisper APIs (Together)
  Together AI's First Voice Offering: High-Performance Transcription at Scale Today marks an important expansion for Together AI.
• Together’s $20M seed funding to build open-source AI and cloud platform (Together)
  We’re excited to announce our $20 M seed round funding round led by Lux Capital. It’s an honor to reach this milestone with the support of…
• FlashFFTConv: Efficient Convolutions for Long Sequences with Tensor Cores (Together)
  Convolution models with long filters now have state-of-the-art reasoning abilities in many long-sequence tasks, from long-range language mo…
• Deploy DeepSeek-R1 at scale: Fast, secure serverless APIs and large-scale Together Reasoning (Together)
  Since the launch of Deep Seek-R1, we've been approached by a surge of companies integrating reasoning models into their production applicat…
• FLUX Tools now available via Together APIs: Get greater control over image generation with Canny (Together)
  Starting today on Together AI, you can access three powerful new image generation models from Black Forest Labs (BFL)'s FLUX Tools suite, g…
• Serving MiniMax-M3 for efficient inference: Unlocking 1M-Token Context and Multimodality Without (Together)
  - Together AI is the preferred cloud partner for Mini Max M3. Together AI will host the open-weights model as a developer endpoint upon its…
• Fine-Tuning Small Open-Source LLMs to Outperform Large Closed-Source Models by 60% on Specialized (Together)
  TL;DR: Parsed, using Together AI’s fine-tuning platform, shows how small open-source models—when paired with rigorous evaluation and task-s…
• Foundational research powering efficient inference at scale (Together)
  AI has spent years in the spotlight for training: the massive, GPU-intensive process of building models.
• Fine-Tuning Platform Upgrades: Larger Models, Longer Contexts, Enhanced Hugging Face Integrations (Together)
  Model customization is an extremely versatile tool that comes in handy for many kinds of AI developers.
• FlashAttention: Fast and memory-efficient exact attention with IO-Awareness (Together)
  Transformers are slow and memory-hungry on long sequences, since the time and memory complexity of self-attention are quadratic in sequence…
• Announcing function calling and JSON mode (Together)
  We are excited to introduce JSON mode & function calling on Together Inference! They are designed to provide you with more flexibility and…
• Speculative decoding for high-throughput long-context inference (Together)
  The amount of inference being performed with LLMs is growing dramatically across many different use cases, many of which utilize the ever-i…
• FlashAttention-4: Algorithm and Kernel Pipelining Co-Design for Asymmetric Hardware Scaling (Together)
  Modern accelerators like Blackwell GPUs continue the trend of asymmetric hardware scaling, where tensor core throughput grows far faster th…
• From 732 bytes to nowhere: shutting down Copy Fail in production (Together)
  We were able to get ahead of Copy Fail (CVE‑2026‑31431) by treating it as a fleet‑level emergency, shutting off the vulnerable crypto socke…
• Building your own RAG application using Together AI and LlamaIndex (Together)
  Together AI provides the fastest cloud platform for building and running generative AI. As part of a series of blog posts about the Togethe…