Together AI

How Together served MiniMax-M3 efficiently with KV-block-major sparse attention, paged MSA decode, optimized index scoring, and a Rust-based multimodal gateway.

High signal Matched: inference, serving

Together AI built the fastest speech-to-text stack on Artificial Analysis by treating ASR as a full-path systems problem, not just a GPU inference problem.

High signal Matched: inference, gpu

Real-world inference benchmarks for coding agents: 31% more TPS than TensorRT-LLM, 2× better TTFT at saturation, and 76% lower cost than Claude Opus 4.6.

High signal Matched: inference, ttft, cost, benchmarks, agents

Together AI partners with Pearl Research Labs to launch a discounted Pearl-powered inference endpoint for Gemma-4-31B-it-pearl, using Proof of Useful Work to turn AI workloads into crypto emissions.

High signal Matched: inference, endpoint, cost, launch, research

Voice finder helps developers search, match, filter, and audition 600+ voices across Together AI TTS models using natural-language prompts or uploaded audio samples.

High signal Matched: introducing

DeepSeek-V4 makes million-token context a serving-systems problem. Together AI explores the inference work behind V4 on NVIDIA HGX B200, including compressed KV layouts, prefix caching, kernel maturity, and endpoint profiles for long-conte...

High signal Matched: inference, serving, endpoint, kernel, b200, long-context

Learn how to deploy any Hugging Face model in one session using Goose and Together's Dedicated Container Inference. Skip the setup complexity — one prompt gets your model running in a production-grade GPU environment on release day.

High signal Matched: inference, gpu, release, model

As AI moves from research to production, the challenge for AI-native teams shifts from building models to running them — efficiently, reliably, and at scale.

High signal Matched: inference, research

DeepSeek-V4 Pro is now available on Together AI with 512K context, controllable reasoning modes, and cached-input pricing for long-context reasoning workloads like code agents, document intelligence, and research synthesis.

High signal Matched: research, long-context, agents

NVIDIA Nemotron 3 Nano Omni is now on Together AI: a single open model that reasons across video, images, audio, and text, built for agentic workloads at scale.

High signal Matched: model, open model, agentic

Rollout is the silent bottleneck in RL post-training. DAS fixes it with adaptive speculative decoding — up to 50% faster, zero degradation in reward quality.

High signal Matched: decoding, speculative decoding, training, post-training

Learn how AI-native companies design multi-tenant GPU clusters that pool capacity without sacrificing team isolation — and how Together AI makes it work in practice.

High signal Matched: gpu

Parcae is a stable looped language model that matches the quality of a Transformer twice its size — a 770M model reaching 1.3B-level performance. We introduce the first scaling laws for looping and show that increasing recurrence, not just...

High signal Matched: performance, model

AI-native companies need infrastructure built for models, not legacy workloads. Learn what defines an AI Native Cloud and why it matters for the next platform shift.

High signal Matched: cloud

A four-model video suite for generation, continuation, reference-driven workflows, and editing, rolling out on Together AI starting with text-to-video.

High signal Matched: generation, model

New research shows LLMs can optimize database query execution plans—achieving up to 4.78x speedups by correcting the cardinality estimation errors that statistical heuristics miss.

High signal Matched: research

Production STT and TTS from Deepgram, available on Together AI Dedicated Model Inference for real-time voice agents.

High signal Matched: inference, model, agents

The team behind FlashAttention and ThunderKittens — how Together AI's kernel researchers close the gap between GPU hardware and production AI.

High signal Matched: kernel, flashattention, gpu

1.25x over a well-trained static speculator. Aurora is an open-source RL framework that turns speculative decoding from a one-time offline setup into a self-improving system that learns from every request it serves.

High signal Matched: decoding, speculative decoding, open-source

As context windows grow, LLM performance degrades in unexpected ways. We show how a "Divide & Conquer" framework — breaking long documents into parallel chunks with a planner, workers, and manager — lets smaller models like Llama-3-70B and...

High signal Matched: performance, long context

Together AI expands fine-tuning with native support for tool call, reasoning, and vision-language models, plus 100B+ model training, up to 6× higher throughput, and job cost and ETA estimates.

High signal Matched: throughput, cost, model, training, fine-tuning

Meet Mamba-3: the SSM built for inference. Faster than Transformers at decode, stronger than Mamba-2, and open-source from day one.

High signal Matched: inference, open-source

Together AI arrives at NVIDIA GTC 2026 with new launches in inference, agents, voice AI, and open models — plus technical sessions from its research and engineering leaders.

High signal Matched: inference, research, agents

Build real-time voice agents on Together AI with co-located STT, LLM, and TTS infrastructure, native Deepgram and Cartesia support, and end-to-end latency under 500ms.

High signal Matched: latency, agents

NVIDIA Nemotron 3 Super is now available on Together AI Dedicated Inference, delivering efficient multi-agent reasoning, a 1M-token context window, and production-grade deployment on managed infrastructure.

High signal Matched: inference, context window, agent

Together GPU Clusters now include built-in autoscaling, RBAC, full-stack observability, and self-healing node repair—giving teams production-ready GPU infrastructure that scales efficiently, stays resilient, and supports shared enterprise...

High signal Matched: gpu

As GPU throughput outpaces memory bandwidth, kernels must evolve. We introduce FlashAttention-4, featuring new pipelining for maximum overlap, 2-CTA MMA modes to reduce shared memory traffic, and a hardware-software hybrid approach to soft...

High signal Matched: throughput, kernel, flashattention, gpu

At AI Native Conf, Together AI announced breakthroughs across kernels, RL, and inference optimization — including FlashAttention-4, ThunderAgent, and together.compile. Research that ships to production. That's the AI Native Cloud.

High signal Matched: inference, flashattention, research, cloud

Serving long prompts doesn't have to mean slow responses. Learn how Together AI's CPD architecture separates warm and cold inference workloads to deliver 40% higher throughput and dramatically lower time-to-first-token for long-context LLM...

High signal Matched: inference, serving, prefill, throughput, long-context

We've refreshed our visual identity — designed with Pentagram to express how Together AI connects open-source innovation, systems research, and builders to unlock new possibilities.

High signal Matched: introducing, research, open-source

State-of-the-art speech models like Whisper and Deepgram score near-human on benchmarks — then fail 39% of the time on street names. New research from Together AI exposes the gap and a fix.

High signal Matched: research, benchmarks

Standard diffusion language models can't use KV caching and need too many refinement steps to be practical. CDLM fixes both with a post-training recipe that enables exact block-wise KV caching and trajectory-consistent step reduction — del...

High signal Matched: inference, latency, training, post-training

Together AI launches production-grade orchestration for custom AI models with 1.4x–2.6x faster inference.

High signal Matched: inference, introducing

What do language models generate when you don't tell them what to generate? New research reveals that LLM families have distinct 'knowledge priors'—GPT models default to code and math, Llama favors narratives, DeepSeek generates religious...

High signal Matched: research

Fine-tuned open-source LLM judges can outperform GPT-5.2 at evaluating model outputs. Using Direct Preference Optimization on just 5,400 preference pairs, we trained GPT-OSS 120B to beat GPT-5.2 on human preference alignment—at 15x lower c...

High signal Matched: inference, cost, model, fine-tuning, evaluating, open-source, oss

Together Evaluations now supports OpenAI, Anthropic, and Google models for cross-provider benchmarking. Compare open-source, fine-tuned, and proprietary models side-by-side to make data-driven decisions on quality, cost, and performance—al...

High signal Matched: performance, cost, open-source, open source

Introducing DSGym—a holisti evaluation and training framework for LLM-based data science agents. Features 90+ bioinformatics tasks, 92 Kaggle competitions, and synthetic trajectory generation. Our 4B model achieves state-of-the-art perform...

High signal Matched: generation, performance, introducing, model, evaluation, training, evaluating, agents, open-source

Learn how to reduce inference latency without massive cost using proven inference optimization tactics — improving throughput, GPU utilization, and cost efficiency while balancing throughput vs. latency tradeoffs.

High signal Matched: inference, throughput, latency, cost, gpu

Together AI teamed with Cursor to build the real-time inference stack that keeps in-editor agents fast and reliable. They productionized NVIDIA Blackwell (B200/GB200), tuning ARM hosts, kernels, and FP4/TensorRT quantization for low latenc...

High signal Matched: inference, latency, b200, gb200, blackwell, model, quantization, agents

Learn how foundation models are trained at scale using multi-node GPU clusters, including distributed training techniques, infrastructure requirements, and practical steps to scale training efficiently.

High signal Matched: distributed, multi-node, gpu, model, training, distributed training

Learn how to choose the right open-source model for production by evaluating model quality, benchmarking performance, and deploying open models that balance cost, speed, and accuracy.

High signal Matched: performance, cost, model, open model, evaluating, open-source

MiniMax Speech 2.6 Turbo: State-of-the-art multilingual TTS with human-level emotional awareness, sub-250ms latency, and 40+ languages—now on Together AI.

High signal Matched: latency

Dan Fu, our VP of Kernels, has published a new post challenging the idea that AI is hitting a hardware wall. He argues that we are vastly underutilizing current chips and that better software-hardware co-design will unlock the next order o...

High signal Matched: performance, research

Nemotron 3 Nano, NVIDIA’s newest reasoning model, is now available on Together AI, the AI Native Cloud

High signal Matched: model, cloud, reasoning model

AutoJudge accelerates LLM inference by identifying which token mismatches actually matter. Using self-supervised learning to train a lightweight classifier, it accepts up to 40 draft tokens per cycle—delivering 1.5–2× speedups over standar...

High signal Matched: inference, decoding, speculative decoding, introducing

Build, train, and deploy advanced AI agents with integrated reinforcement learning on the Together platform.

High signal Matched: cloud, agents

No feed summary available yet.

High signal Matched: cloud

Together AI achieves up to 2x faster inference for top open-source models like Qwen, DeepSeek, and Kimi through GPU optimization, advanced speculative decoding, and FP4 quantization—ranking #1 in speed benchmarks on NVIDIA Blackwell archit...

High signal Matched: inference, decoding, speculative decoding, gpu, blackwell, quantization, benchmarks, open-source

Production-grade image generation with multi-reference consistency, exact brand colors, and reliable text rendering. FLUX.2 from Black Forest Labs, now on Together AI's platform.

High signal Matched: generation

Together AI launches the fastest voice AI stack: streaming Whisper STT, serverless open-source TTS (Orpheus & Kokoro), and Voxtral transcription. Sub-second latency for production voice agents.

High signal Matched: inference, latency, agents, open-source

Understanding how to evaluate and benchmark Large Language Models (LLMS). Test, compare, and understand LLMs.

High signal Matched: benchmark, evaluate

ReasonIF finds frontier LRMs fail to follow reasoning instructions >75% of the time; introduces a benchmark across languages, formatting, and length.

High signal Matched: benchmark

Together AI adds 40+ image & video models, including Sora 2 and Veo 3, to build end-to-end multimodal apps with unified OpenAI-compatible APIs and transparent pricing.

High signal Matched: generation, model, openai-compatible

We've launched the Together AI Startup Accelerator: Up to $50K credits, expert engineering hours, GTM support, community and VC access for AI-native apps in build–scale tiers.

High signal Matched: accelerator

LLM inference that gets faster as you use it. Our runtime-learning accelerator adapts continuously to your workload, delivering 500 TPS on DeepSeek-V3.1, a 4x speedup over baseline performance without manual tuning.

High signal Matched: inference, deepseek-v3, performance, accelerator

Our new Batch Inference API makes large-scale AI workloads simpler, faster, and cheaper. With a streamlined UI, universal model support, and 3000× higher rate limits—now up to 30B tokens—you can process massive datasets at half the cost of...

High signal Matched: inference, cost, model, api

Together AI launches Instant Clusters: self-service GPU clusters with NVIDIA H100/B200, ready in minutes for training or inference at any scale.

High signal Matched: inference, gpu, h100, b200, training

Access DeepSeek-V3.1 on Together AI: MIT-licensed hybrid model with thinking/non-thinking modes, 66% SWE-bench Verified, serverless deployment, 99.9% SLA.

High signal Matched: deepseek-v3, model, swe-bench

Build AI agents for complex, long-running engineering tasks. Learn key patterns from a case study: accelerating LLM inference with speculative decoding.

High signal Matched: inference, decoding, speculative decoding, agents

Customize OpenAI’s gpt-oss-20B/120B with Together AI’s fine-tuning: train, optimize, and instantly deploy domain experts with enterprise reliability and cost efficiency.

High signal Matched: cost, fine-tuning, oss

Parsed fine-tuned a 27B open-source model to beat Claude Sonnet 4 by 60% on a real-world healthcare task—while running 10–100x cheaper.

High signal Matched: model, fine-tuning, open-source

Access OpenAI’s gpt-oss-120B on Together AI: Apache-2.0 open-weight model with serverless & dedicated endpoints, $0.50/1M in, $1.50/1M out, 99.9% SLA.

High signal Matched: model, oss

Together Evaluations is a flexible framework for benchmarking LLMs using strong open-source models as judges. Skip manual labeling and rigid metrics—get fast, customizable insights into model quality for your specific tasks.

High signal Matched: benchmark, model, open-source

Unlock agentic coding with Qwen3-Coder on Together AI: 256K context, SWE-bench rivaling Claude Sonnet 4, zero-setup instant deployment.

High signal Matched: model, swe-bench, agentic

Together AI inference is now among the world’s fastest, most capable platforms for running open-source reasoning models like DeepSeek-R1 at scale, thanks to our new inference engine designed for NVIDIA HGX B200.

High signal Matched: inference, b200, blackwell, open-source

Run Kimi K2 (1T params) on Together AI—frontier open model for agentic reasoning and coding, serverless deployment, 99.9% SLA, lower cost and instant scaling.

High signal Matched: cost, model, open model, agentic, open-source

No feed summary available yet.

High signal Matched: performance

No feed summary available yet.

High signal Matched: cost, introducing, api

No feed summary available yet.

High signal Matched: model

No feed summary available yet.

High signal Matched: introducing, sota

No feed summary available yet.

High signal Matched: decoding, speculative decoding

No feed summary available yet.

High signal Matched: inference

No feed summary available yet.

High signal Matched: blackwell

Violin is an open-source AI video translation tool that combines speech recognition, LLM translation, and text-to-speech to make video content accessible across languages.

Watchlist Matched: open-source

No feed summary available yet.

Watchlist Matched: none

Together AI and Adaption partner to bring Together Fine-Tuning natively into Adaptive Data, helping teams optimize datasets, run fine-tuning, evaluate results, and deploy stronger open models.

Watchlist Matched: fine-tuning, evaluate

EinsteinArena is a platform where AI agents collaborate and compete on open math problems. AI agents on EinsteinArena have already set 11 new state-of-the-art results on open math problems — including pushing the kissing number lower bound...

Watchlist Matched: agents

No feed summary available yet.

Watchlist Matched: training, agents, sota

No feed summary available yet.

Watchlist Matched: none

Hiring Alon Gavrielov further deepens Together AI’s commitment to building AI factories that deliver the most reliable, efficient, and scalable infrastructure for AI-native teams.

Watchlist Matched: none

Two enterprise-grade Rime TTS models now available on Together AI. Co-locate with LLM and STT on dedicated infrastructure. Proven at billions of calls.

Watchlist Matched: none

No feed summary available yet.

Watchlist Matched: sdk

Test AI agents in the real world with Collinear TraitMix and Together Evals: dynamic persona simulations, multi-turn dialogs, and LLM-as-judge scoring.

Watchlist Matched: evals, agent, agents

Together AI expands Fine-Tuning Platform: train 100B+ models, extend context lengths, integrate with Hugging Face Hub, and access new DPO options.

Watchlist Matched: dpo, fine-tuning

Hiring Mahadev Konar further deepens Together AI’s commitment to deliver the most reliable and scalable GPU infrastructure.

Watchlist Matched: gpu

No feed summary available yet.

Watchlist Matched: oss

No feed summary available yet.

Watchlist Matched: none

Build and deploy AI with peace of mind—Together AI is now SOC 2 Type 2 certified, proving our encryption, access controls, and 24/7 monitoring meet the highest security standards.

Watchlist Matched: none

No feed summary available yet.

Watchlist Matched: training, agent

No feed summary available yet.

Watchlist Matched: none

Build a data scientist agent using Together’s open-source models and Code Interpreter—easy to implement, solid benchmarks, and full code on GitHub.

Watchlist Matched: benchmarks, agent, open-source

No feed summary available yet.

Watchlist Matched: none

No feed summary available yet.

Watchlist Matched: fine-tuning

No feed summary available yet.

Watchlist Matched: training, post-training, agents, open-source

No feed summary available yet.

Watchlist Matched: none

No feed summary available yet.

Watchlist Matched: api

No feed summary available yet.

Watchlist Matched: none

No feed summary available yet.

Watchlist Matched: training

No feed summary available yet.

Watchlist Matched: training, fine-tuning