training

hardware model-release cloud training

High signal Matched: performance, model, training, checkpointing, fine-tuning

Lambda · cloud · 2026-06-03

Introducing workspaces for Lambda Cloud

Score 17

Lambda workspaces help teams organize cloud resources, control access, and separate dev, staging, and production in shared GPU environments. A junior researcher kills a production training run. A contractor sees weights they shouldn't. If...

inference training quantization

High signal Matched: gpu, introducing, weights, cloud, training

vLLM Project · open-source · 2026-06-02

Accelerating vLLM-Omni Inference with AutoRound Quantization

Score 13

We are excited to announce that AutoRound — Intel's state-of-the-art post-training quantization (PTQ) algorithm — is now fully integrated into vLLM-Omni, enabling a streamlined quantize-once,...

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

NVIDIA Technical Blog · hardware · 2026-06-01

NVIDIA Vera CPU Sets a New Standard for Agentic Workloads in AI Factories

Score 11

Each wave of AI has created a new scaling law. Pretraining scaled intelligence through larger datasets, more parameters, and massively parallel GPU systems....

hardware training agents

model-release cloud training

High signal Matched: gpu, pretraining, agentic

AWS Machine Learning Blog · cloud · 2026-05-29

Training Azerbaijani language models on Amazon SageMaker AI

Score 13

Azercell Telecom LLC, Azerbaijan's leading telecommunications provider, wanted to build an Azerbaijani large language model (LLM) on Amazon SageMaker AI for telecom use cases and a customer-facing chatbot. The challenge: adapting foundatio...

inference speculative-decoding model-release training

High signal Matched: model, sagemaker, training

vLLM Project · open-source · 2026-05-28

Speculators v0.5.0: DFlash Support and Online Training

Score 19

The v0.5.0 release brings significant architectural improvements to speculative decoding model training, introducing DFlash algorithm support, fully unified online training capabilities, and a...

High signal Matched: decoding, speculative decoding, release, introducing, model, training

vLLM Project · open-source · 2026-05-28

Native RL APIs in vLLM

Score 11

As post-training workloads continue to scale, we've seen widespread adoption of vLLM as the inference engine of choice. However, two issues repeatedly arise:

inference training

High signal Matched: inference, training, post-training

Lambda · cloud · 2026-05-21

Lambda Bare Metal Instances: full hardware control with API-driven operations

Score 8

The unit of AI compute has shifted from single hosts to rack-scale systems that integrate NVIDIA GPUs, CPUs, scale-up networking fabrics, and liquid cooling, such as the NVIDIA GB300 NVL72 and NVIDIA Vera Rubin NVL72. Teams at the frontier...

inference serving benchmark cloud training api

High signal Matched: serving, performance, cloud, training, api

vLLM Project · open-source · 2026-05-14

Announcing VeRL-Omni: Easy, Fast, and Stable RL Training for Diffusion and Omni-Modality Models

Score 10

We are excited to announce the pre-release of VeRL-Omni, a general reinforcement learning (RL) post-training framework focused on multimodal generative models, built on top of verl and vllm-omni.

inference model-release training

High signal Matched: release, training, post-training

Hugging Face · open-source · 2026-05-12

Building Blocks for Foundation Model Training and Inference on AWS

Score 14

No feed summary available yet.

inference serving kv-cache speculative-decoding benchmark model-release research training fine-tuning evals long-context agents frontier-model

High signal Matched: inference, model, training

BAIR · research · 2026-05-08

Adaptive Parallel Reasoning: The Next Paradigm in Efficient Inference Scaling

Score 28

.apr-fig { text-align: center; margin: 1.35em 0; line-height: 1.4; } .apr-fig--wide img { display: inline-block; width: 100%; max-width: 100%; height: auto; vertical-align: middle; } .apr-fig--wide-0-8 { max-width: 80%; margin-left: auto;...

moe benchmark model-release training

High signal Matched: inference, decoding, prefill, generation, serve, throughput, kv cache, verification, performance, latency, cost, model, paper, research, evaluation, training, pretraining, sft, benchmarks, long context, context window, agentic, reasoning model

AI2 · research · 2026-05-08

EMO: Pretraining mixture of experts for emergent modularity

Score 12

EMO is a new mixture-of-experts model trained so modular expert groups emerge from data, enabling users to select small task-specific expert subsets while preserving near full-model performance.

inference benchmark model-release training quantization

High signal Matched: mixture of experts, performance, model, pretraining

NVIDIA Technical Blog · hardware · 2026-05-07

Model Quantization: Post-Training Quantization Using NVIDIA Model Optimizer

Score 16

Model quantization is an effective method to reduce VRAM usage and improve inference performance on consumer devices such as NVIDIA GeForce RTX GPUs. By...

distributed benchmark hardware training

High signal Matched: inference, performance, model, training, post-training, quantization

NVIDIA Technical Blog · hardware · 2026-05-07

Real-Time Performance Monitoring and Faster Debugging with NCCL Inspector and Prometheus

Score 20

Distributed deep learning depends on fast, reliable GPU-to-GPU communication using the NVIDIA Collective Communication Library (NCCL). When training slows down,...

distributed training evals

High signal Matched: distributed, nccl, performance, gpu, training

SkyPilot · open-source · 2026-05-01

Cache Me If You Can: Tuning Object Stores for AI

Score 8

We ran hundreds of benchmarks to tune storage systems for distributed training so you don’t have to.

High signal Matched: distributed, training, distributed training, benchmarks

Nota AI · korea · 2026-04-29

[NVIDIA Nemotron Hackathon] Grand Prize Among 20 Teams: Behind Two Sleepless Days

Score 32

  Hancheol Park, Ph. D.AI Research Engineer, NetsPresso Tech, Nota AI Geonmin Kim, Ph. D.AI Research Engineer, NetsPresso Tech, Nota AI Geonho LeeEdge AI Engineer Intern, NetsPresso Tech, Nota AI Jaehoon Lee Technical Content Manager,...

inference moe benchmark model-release research korea training fine-tuning quantization evals agents

inference speculative-decoding training

High signal Matched: generation, moe, performance, model, weights, paper, research, evaluation, korea, korean, seoul, naver, training, fine-tuning, quantization, agent, agents, agentic

Together AI · inference-infra · 2026-04-24

Accelerate RL rollouts by up to 50% with distribution-aware speculative decoding

Score 16

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

Nota AI · korea · 2026-04-22

[Deep Dive: NetsPresso®] From Quantization to Graph Optimization: A Step-by-Step Model Deployment Pipeline

Score 54

  Jaehoon Lee Technical Content Manager, Nota AI   Series Notice: NetsPresso® Technical Blog, Part 2In Part 1, we walked through a scenario of deploying Llama 3.2 1B on an edge device to illustrate the NetsPresso® workflow. The f...

inference kernel cuda benchmark hardware model-release research korea training quantization evals api open-source

inference serving benchmark model-release training quantization

High signal Matched: inference, kernel, cuda, matmul, benchmark, performance, latency, cost, npu, model, weights, paper, research, evaluation, furiosa, training, quantization, int8, int4, awq, gptq, sdk, open-source

NVIDIA Technical Blog · hardware · 2026-04-20

Run High-Throughput Reinforcement Learning Training with End-to-End FP8 Precision

Score 18

As LLMs transition from simple text generation to complex reasoning, reinforcement learning (RL) plays a central role. Algorithms like Group Relative Policy...

benchmark model-release research training evals

High signal Matched: generation, throughput, fp8, training

BAIR · research · 2026-04-20

Gradient-based Planning for World Models at Longer Horizons

Score 16

.grasp-results-table table { font-size: 0.875rem; line-height: 1.35; width: 100%; } .grasp-results-table th, .grasp-results-table td { padding: 0.35rem 0.5rem; } /* Consistent whitespace between major sections (this post is long and hr-hea...

model-release cloud training agents

High signal Matched: performance, model, paper, arxiv, evaluation, training

SkyPilot · open-source · 2026-04-10

SkyPilot Agent Skill: Let Agents Manage Your GPUs

Score 10

With the SkyPilot Agent Skill, your AI coding agent can launch clusters, run training jobs and manage cloud resources across any infrastructure using natural language.

High signal Matched: launch, cloud, training, agent, agents

NVIDIA Technical Blog · hardware · 2026-04-09

Cut Checkpoint Costs with About 30 Lines of Python and NVIDIA nvCOMP

Score 16

Training LLMs requires periodic checkpoints. These full snapshots of model weights, optimizer states, and gradients are saved to storage so training can resume...

inference serving kv-cache benchmark hardware model-release research training fine-tuning quantization agents frontier-model

High signal Matched: model, weights, checkpoint, training

Nota AI · korea · 2026-03-31

The Real Reason TurboQuant Shook the Market: AI Optimization Has Gone Mainstream

Score 46

  Jaehoon Lee Technical Content Manager, Nota AI   In March, a single official announcement from Google Research rocked trillions of won in the market capitalization of U.S. infrastructure and semiconductor stocks. The catalyst:...

High signal Matched: inference, serving, generation, throughput, kv cache, benchmark, performance, cost, b200, blackwell, introducing, model, fp8, research, training, fine-tuning, quantization, quantized, agent, agentic, frontier model

Nota AI · korea · 2026-03-23

[GTC 2026 Recap] The Trillion-Dollar Inference Race Begins: How Nota AI Fills the Gap

Score 42

  Jaehoon Lee Technical Content Manager, Nota AI   GTC has evolved far beyond a technology conference, drawing attention from global economies and financial markets alike. This year, CEO Jensen Huang took the stage in his tradema...

inference serving kernel cuda kv-cache benchmark hardware model-release research cloud training long-context agents open-source

serving benchmark model-release training fine-tuning

High signal Matched: inference, prefill, generation, throughput, cuda, kv cache, performance, latency, cost, gpu, npu, launch, model, research, cloud, training, long-context, context window, agent, agents, agentic, open-source

Together AI · inference-infra · 2026-03-18

Together AI expands fine-tuning service with tool calling, reasoning, and vision support

Score 14

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.

inference serving moe benchmark hardware model-release research korea training quantization evals long-context open-source

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

Nota AI · korea · 2026-03-13

NotaMoEQuantization: An MoE-Specific Quantization Method for Solar-Open-100B

Score 62

  Hancheol Park, Ph. D. AI Research Engineer, Nota AI Tairen PiaoAI Research Engineer, Nota AI Tae-Ho KimCTO & Co-Founder, Nota AI ✔️ Resource : The official quantized model of Solar-Open-100B, which passed the first round of Sout...

inference serving benchmark model-release research training evals long-context rag

High signal Matched: inference, serving, prefill, generation, throughput, moe, router, benchmark, performance, latency, ttft, tpot, blackwell, release, model, weights, open model, research, evaluation, korea, korean, upstage, training, post-training, quantization, quantized, int4, evaluate, benchmarks, mmlu, long-context

BAIR · research · 2026-03-13

Identifying Interactions at Scale for LLMs

Score 18

--> Understanding the behavior of complex machine learning systems, particularly Large Language Models (LLMs), is a critical challenge in modern artificial intelligence. Interpretability research aims to make the decision-making process mo...

High signal Matched: inference, serving, decoding, performance, cost, model, research, training, evaluate, mmlu, long-context, rag

NVIDIA Technical Blog · hardware · 2026-03-13

Scale Synthetic Data and Physical AI Reasoning with NVIDIA Cosmos World Foundation Models

Score 10

The next generation of AI-driven robots like humanoids and autonomous vehicles depends on high-fidelity, physics-aware training data. Without diverse and...

inference training

High signal Matched: generation, training

Hugging Face · open-source · 2026-03-04

PRX Part 3 — Training a Text-to-Image Model in 24h!

Score 10

No feed summary available yet.

inference speculative-decoding benchmark model-release research training evals

High signal Matched: model, training

Nota AI · korea · 2026-02-26

ERGO: Efficient High-Resolution Visual Understanding for Vision-Language Models

Score 24

inference benchmark training

High signal Matched: inference, generation, verification, benchmark, performance, latency, cost, model, arxiv, evaluation, training, post-training, benchmarks

Together AI · inference-infra · 2026-02-19

Consistency diffusion language models: Up to 14x faster inference without sacrificing quality

Score 14

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...

inference benchmark model-release research training evals agents open-source

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

Together AI · inference-infra · 2026-01-26

DSGym: A holistic framework for evaluating and training data science agents

Score 18

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...

distributed hardware model-release training

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

Together AI · inference-infra · 2026-01-12

Inside multi-node training: How to scale model training across GPU clusters

Score 22

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.

benchmark model-release research training evals

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

BAIR · research · 2026-01-10

Information-Driven Design of Imaging Systems

Score 12

An encoder (optical system) maps objects to noiseless images, which noise corrupts into measurements. Our information estimator uses only these noisy measurements and a noise model to quantify how well measurements distinguish objects. Man...

High signal Matched: performance, model, paper, evaluation, training, evaluate

Nota AI · korea · 2025-12-19

NVIDIA Blackwell; The Impact of NVFP4 For LLM Inference

Score 74

  Seungmin YangEdgeFM Lead, Nota AI On this page ▾ SummaryWith the introduction of NVFP4—a new 4-bit floating point data type in NVIDIA’s Blackwell GPU architecture—LLM inference achieves markedly improved efficiency.Blackwell’s NVFP4...

inference serving kernel cuda distributed benchmark hardware model-release research training quantization evals rag

inference speculative-decoding benchmark model-release training

High signal Matched: inference, serving, decoding, prefill, generation, token generation, throughput, kernel, gemm, cutlass, distributed, benchmark, performance, latency, ttft, tpot, tokens/sec, cost, gpu, blackwell, launch, model, weights, fp8, research, training, post-training, quantization, quantized, awq, benchmarks, mmlu, retrieval

vLLM Project · open-source · 2025-12-13

Diving into speculative decoding training support for vLLM with Speculators v0.3.0

Score 24

- Speculative decoding serves as an optimization to improve inference performance; however, training a unique draft model for each LLM can be difficult and time-consuming, while production-ready...

inference serving kernel benchmark hardware training

High signal Matched: inference, decoding, speculative decoding, draft model, performance, model, training

SqueezeBits · korea · 2025-10-28

[Intel Gaudi] #6. GEMM, Attention, vLLM on Gaudi

Score 20

Explore how Intel’s new Gaudi-3 compares to Gaudi-2, NVIDIA A100, and H100. We analyze real-world GEMM efficiency, attention performance, and LLM serving results to uncover what truly matters for AI inference and training workloads.

High signal Matched: inference, serving, gemm, performance, h100, training

SkyPilot · open-source · 2025-09-11

From 1 hour to 10 minutes: How I sped up my distributed LLM training without changing the code or GPUs

Score 10

This page has moved. If you are not redirected automatically, click here.

distributed training

inference hardware training

High signal Matched: distributed, training

Together AI · inference-infra · 2025-09-09

Announcing General Availability of Together Instant Clusters, offering ready to use, self-service NVIDIA GPUs

Score 18

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

benchmark model-release research training

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

BAIR · research · 2025-09-01

What exactly does word2vec learn?

Score 14

What exactly does word2vec learn, and how? Answering this question amounts to understanding representation learning in a minimal yet interesting language modeling task. Despite the fact that word2vec is a well-known precursor to modern lan...

distributed hardware training

High signal Matched: benchmark, performance, model, weights, paper, training

Hugging Face · open-source · 2025-08-08

Accelerate ND-Parallel: A guide to Efficient Multi-GPU Training

Score 14

No feed summary available yet.

distributed hardware model-release training

High signal Matched: multi-gpu, gpu, training

Modal · inference-infra · 2025-07-11

Product updates: Multi-node training clusters, B200 and H200s, and Client 1.0 release

Score 18

Welcome to another round of Modal Product Updates! Here's what's new this month.

inference benchmark model-release research training fine-tuning evals

High signal Matched: multi-node, b200, release, training

Nota AI · korea · 2025-07-10

Video Self-Distillation for Single-Image Encoders: Learning Temporal Priors from Unlabeled Video

Score 20

  Marcel Simon, Ph. D.ML Researcher, Nota AI GmbH Tae-Ho KimCTO & Co-Founder, Nota AI Seul-Ki Yeom, Ph. D.Research Lead, Nota AI GmbH   SummaryProposes a simple next-frame prediction task using unlabeled video to enhance sing...

High signal Matched: inference, performance, model, paper, research, training, fine-tuning, benchmarks

Hugging Face · open-source · 2025-07-04

Announcing NeurIPS 2025 E2LM Competition: Early Training Evaluation of Language Models

Score 10

No feed summary available yet.

research training evals

inference benchmark model-release research training evals agents

High signal Matched: evaluation, training

BAIR · research · 2025-07-01

Whole-Body Conditioned Egocentric Video Prediction

Score 10

.modal { display: none; position: fixed; z-index: 9999; padding-top: 50px; left: 0; top: 0; width: 100%; height: 100%; overflow: auto; background-color: rgba(0,0,0,0.9); } .modal-content { margin: auto; display: block; max-width: 90%; max-...

High signal Matched: inference, generation, performance, model, paper, arxiv, evaluation, training, evaluate, agent, agents

Hugging Face · open-source · 2025-06-11

Introducing Training Cluster as a Service - a new collaboration with NVIDIA

Score 10

No feed summary available yet.

inference kv-cache benchmark model-release research training evals open-source

High signal Matched: introducing, training

Nota AI · korea · 2025-05-07

Efficient LLaMA-3.2-Vision by Trimming Cross-attended Visual Features</span#x3E;

Score 28

High signal Matched: inference, generation, kv cache, benchmark, performance, latency, model, weights, research, training, benchmarks, open-source

Modal · inference-infra · 2025-04-18

How sync. uses Modal to lipsync 100 hours of video a day

Score 8

sync. is a research lab training foundational models to understand and manipulate humans in video. After outgrowing Google Colab, they partnered with Modal for efficient deployment, allowing rapid iteration and scaling to process over 100...

research training

benchmark model-release research training fine-tuning evals rag api frontier-model

High signal Matched: research, training

BAIR · research · 2025-04-11

Defending against Prompt Injection with Structured Queries (StruQ) and Preference Optimization (SecAlign)

Score 10

Recent advances in Large Language Models (LLMs) enable exciting LLM-integrated applications. However, as LLMs have improved, so have the attacks against them. Prompt injection attack is listed as the #1 threat by OWASP to LLM-integrated ap...

inference benchmark model-release research training rag

High signal Matched: cost, model, evaluation, training, dpo, fine-tuning, retrieval, api, sota

BAIR · research · 2025-04-08

Repurposing Protein Folding Models for Generation with Latent Diffusion

Score 20

PLAID is a multimodal generative model that simultaneously generates protein 1D sequence and 3D structure, by learning the latent space of protein folding models. The awarding of the 2024 Nobel Prize to AlphaFold2 marks an important moment...

benchmark model-release cloud training

High signal Matched: inference, generation, cost, model, weights, research, training, retrieval

SkyPilot · open-source · 2025-04-08

High-Performance Model Checkpointing on the Cloud

Score 18

Techniques to speed up checkpointing by 9.6x and how to easily achieve them in SkyPilot

inference distributed benchmark model-release training long-context

High signal Matched: performance, model, cloud, checkpointing

AIBrix · open-source · 2025-03-10

DeepSeek-R1 671B multi-host Deployment in AIBrix

Score 20

This blog post introduces deploying DeepSeek R1 using AIBrix. DeepSeek-R1 demonstrates remarkable proficiency in reasoning tasks through step-by-step training process. It features 671B total parameters with 37B active parameters, and 128k...

inference benchmark model-release research training fine-tuning

High signal Matched: inference, distributed, benchmark, model, weights, training, context length

Nota AI · korea · 2025-02-25

A Study on Detecting LLM-Generated Multilingual Content

Score 18

  Hancheol Park, Ph. D.AI Research Engineer, Nota AI Geonmin Kim, Ph. D.AI Research Engineer, Nota AI Jaeyeon KimAI Research Engineer, Nota AI   SummaryIn this study, we propose a method for determining whether given multilingual...

benchmark research training evals

High signal Matched: generation, performance, model, paper, research, training, fine-tuning

Nota AI · korea · 2025-02-10

Where do LLMs Encode the Knowledge to Assess the Ambiguity?

Score 16

  Hancheol Park, Ph. D.AI Research Engineer, Nota AI Geonmin Kim, Ph. D.AI Research Engineer, Nota AI   SummaryIn this study, we present a method for detecting ambiguous samples in natural language understanding (NLU) tasks using...

inference benchmark model-release research cloud training fine-tuning evals open-source

High signal Matched: performance, paper, research, evaluation, training, evaluate

Nota AI · korea · 2024-08-02

Deploying an Efficient Vision-Language Model on Mobile Devices

Score 38

  Jaeyeon KimResearch Engineer, Nota AI Geonmin KimResearch Engineer, Nota AI Hancheol ParkTeam Lead of NetsPresso Application, Nota AI   IntroductionRecent large language models (LLMs) have demonstrated unprecedented performance...

High signal Matched: decoding, benchmark, performance, latency, tokens/sec, model, arxiv, research, technical report, evaluation, cloud, training, lora, benchmarks, leaderboard, open-source

Replicate · inference-infra · 2024-06-12

H100s are coming to Replicate

Score 8

We'll soon support NVIDIA's H100 GPUs for predictions and training. Let us know if you want early access.

hardware training

High signal Matched: h100, training

Modal · inference-infra · 2024-05-20

Why Substack moved their AI and ML pipelines to Modal

Score 8

Learn how Substack sped up their developer iteration cycles by moving ML training and deployment to Modal from AWS SageMaker.

cloud training

High signal Matched: sagemaker, training

Hugging Face · open-source · 2024-03-20

GaLore: Advancing Large Model Training on Consumer-grade Hardware

Score 10

No feed summary available yet.

High signal Matched: model, training

Replicate · inference-infra · 2023-05-26

Make any large language model a better poet

Score 8

Prompt engineering and training are often the first solutions we reach for to improve language model behavior, but they're not the only way.

High signal Matched: model, training

Hugging Face · open-source · 2023-04-27

Training a language model with 🤗 Transformers using TensorFlow and TPUs

Score 10

No feed summary available yet.

hardware training fine-tuning

High signal Matched: model, training

Hugging Face · open-source · 2023-03-09

Fine-tuning 20B LLMs with RLHF on a 24GB consumer GPU

Score 10

No feed summary available yet.

High signal Matched: gpu, rlhf, fine-tuning

Hugging Face · open-source · 2022-12-14

Faster Training and Inference: Habana Gaudi®2 vs Nvidia A100 80GB

Score 10

No feed summary available yet.

inference training

High signal Matched: inference, training

Hugging Face · open-source · 2022-10-21

From PyTorch DDP to Accelerate to Trainer, mastery of distributed training with ease

Score 10

No feed summary available yet.

distributed training

High signal Matched: distributed, training, distributed training

Hugging Face · open-source · 2022-06-28

Accelerate Large Model Training using DeepSpeed

Score 10

No feed summary available yet.

High signal Matched: model, training

Hugging Face · open-source · 2022-05-02

Accelerate Large Model Training using PyTorch Fully Sharded Data Parallel

Score 10

No feed summary available yet.

High signal Matched: model, training

Hugging Face · open-source · 2022-04-12

Habana Labs and Hugging Face Partner to Accelerate Transformer Model Training

Score 10

No feed summary available yet.

High signal Matched: model, training

Hugging Face · open-source · 2021-10-25

Train a Sentence Embedding Model with 1B Training Pairs

Score 10

No feed summary available yet.

distributed cloud training

High signal Matched: model, training

Hugging Face · open-source · 2021-04-08

Distributed Training: Train BART/T5 for Summarization using 🤗 Transformers and Amazon SageMaker

Score 14

No feed summary available yet.

High signal Matched: distributed, sagemaker, training, distributed training

Prime Intellect · inference-infra · 2026-06-03

Start training

Score 6

No feed summary available yet.

Watchlist Matched: training

Prime Intellect · inference-infra · 2026-06-03

announcementsReleasing Lab: the training platform for self-improving agents

Score 6

No feed summary available yet.

Watchlist Matched: training, agents

Fireworks AI · inference-infra · 2026-06-03

Score 6

No feed summary available yet.

Watchlist Matched: training

Fireworks AI · inference-infra · 2026-06-03

Notes on DeepSeek-V4's training system

Score 6

No feed summary available yet.

Watchlist Matched: training

Anyscale · inference-infra · 2026-06-03

Ray Training

Score 4

No feed summary available yet.

Watchlist Matched: training

Fireworks AI · inference-infra · 2026-06-03

Own Your AI: Fireworks Training Preview

Score 0

No feed summary available yet.

Watchlist Matched: training

NVIDIA Technical Blog · hardware · 2026-06-01

How to Post-Train Autonomous Vehicle Models in Closed-Loop with NVIDIA Alpamayo

Score 4

Developing autonomous vehicle (AV) policies requires bridging an important gap between training and deployment. Vision-language-action (VLA) models that can...

Watchlist Matched: training

Sakana AI · model-lab · 2026-05-28

DiffusionBlocks: Training Neural Networks One Block at a Time

Score 1

No feed summary available yet.

benchmark model-release training

Watchlist Matched: training

Lambda · cloud · 2026-05-04

Most AI teams treat compute as a commodity. It's not.

Score 6

Consider two teams provisioning 8,192 GPUs for a large training run. Same model, same dataset, same budget. Team A lands on a facility purpose-built for AI with sufficient power density, carefully engineered liquid cooling, a high-performa...

hardware training agents open-source

Watchlist Matched: performance, model, training

Lambda · cloud · 2026-04-30

Creating highly efficient agents: 450M tool-calling tokens distilled for post-training from top open-source models

Score 4

Harnesses If you've used Claude Code or Codex, you've used a harness. A harness is the infrastructure layer that wraps an AI coding agent and decides how it operates, what it can touch, and how you measure whether it worked. It's how most...

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

AI2 · research · 2026-04-23

OlmPool: How small architectural choices compound to undermine long context extension

Score 0

OlmPool is a controlled suite of 26 models showing how small architecture choices can compound to make long-context extension much harder, even when training data and extension recipes are held constant.

training long-context

Watchlist Matched: training, long context, long-context

NVIDIA Technical Blog · hardware · 2026-04-22

Advancing Emerging Optimizers for Accelerated LLM Training with NVIDIA Megatron

Score 3

Higher-order optimization algorithms such as Shampoo have been effectively applied in neural network training for at least a decade. These methods have achieved...

Watchlist Matched: training

AI2 · research · 2026-04-20

Train separately, merge together: Modular post-training with mixture-of-experts

Score 6

BAR is a recipe for post-training language models one capability at a time—train domain experts independently, merge them into a single mixture-of-experts model, and upgrade any expert without impacting the others.

Watchlist Matched: model, training, post-training

Hugging Face · open-source · 2026-04-16

Training and Finetuning Multimodal Embedding & Reranker Models with Sentence Transformers

Score 1

No feed summary available yet.

Watchlist Matched: training, finetuning

Hugging Face · open-source · 2026-03-31

Training mRNA Language Models Across 25 Species for $165

Score 1

No feed summary available yet.

Watchlist Matched: training

Hugging Face · open-source · 2026-03-31

TRL v1.0: Post-Training Library Built to Move with the Field

Score 1

No feed summary available yet.

model-release training agents

Watchlist Matched: training, post-training

AI2 · research · 2026-03-24

MolmoWeb: An open agent for automating web tasks

Score 6

Introducing MolmoWeb, an open visual web agent that navigates and completes tasks in a browser using screenshots alone, along with MolmoWebMix, the largest public dataset for training web agents.

model-release training fine-tuning

Watchlist Matched: introducing, training, agent, agents

AI2 · research · 2026-03-11

MolmoBot: Training robot manipulation entirely in simulation

Score 6

MolmoBot is an open robotic manipulation model suite trained entirely in simulation—demonstrating zero-shot transfer to real-world robots without any real-world data collection or fine-tuning.

Watchlist Matched: model, training, fine-tuning

AI2 · research · 2026-03-11

Ai2 introduces open, simulation-first stack for physical AI, achieving zero-shot transfer to real robots

Score 6

Introducing MolmoBot and MolmoSpaces, an open foundation for training real-world robots to advance science.

Watchlist Matched: introducing, training

Hugging Face · open-source · 2026-03-09

Ulysses Sequence Parallelism: Training with Million-Token Contexts

Score 1

No feed summary available yet.

training agents frontier-model

Watchlist Matched: training

Together AI · inference-infra · 2026-02-25

CoderForge-Preview: SOTA open dataset for training efficient coding agents

Score 3

No feed summary available yet.

Watchlist Matched: training, agents, sota

Hugging Face · open-source · 2026-02-03

Training Design for Text-to-Image Models: Lessons from Ablations

Score 1

No feed summary available yet.

training agents open-source

Watchlist Matched: training

Hugging Face · open-source · 2026-01-27

Unlocking Agentic RL Training for GPT-OSS: A Practical Retrospective

Score 1

No feed summary available yet.

benchmark model-release research training

Watchlist Matched: training, agentic, oss

BAIR · research · 2025-11-01

RL without TD learning

Score 4

In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer. Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalabilit...

Watchlist Matched: benchmark, performance, model, paper, training

SkyPilot · open-source · 2025-10-14

How to train and scale AI math/coding agents using VeRL on any AI infra

Score 1

Want to train an AI agent with RL that can solve math problems or write code? This tutorial walks you through building your own math and coding agents with step-by-step examples with plenty of screenshots to help you along the way. We use...

Watchlist Matched: training, post-training, agent, agents

Hugging Face · open-source · 2025-09-23

Smol2Operator: Post-Training GUI Agents for Computer Use

Score 1

No feed summary available yet.

Watchlist Matched: training, post-training, agents, computer use

Together AI · inference-infra · 2025-09-10

Fine-Tuning Platform Upgrades: Larger Models, Longer Contexts, Enhanced Hugging Face Integrations

Score 3

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

Hugging Face · open-source · 2025-09-10

Jupyter Agents: training LLMs to reason with notebooks

Score 1

No feed summary available yet.

Watchlist Matched: training, agents

Google Research · big-tech · 2025-08-07

Achieving 10,000x training data reduction with high-fidelity labels

Score 0

Human-Computer Interaction and Visualization

Watchlist Matched: training

Together AI · inference-infra · 2025-07-02

DeepSWE: Training a Fully Open-sourced, State-of-the-Art Coding Agent by Scaling RL

Score 3

No feed summary available yet.

Watchlist Matched: training, agent

Hugging Face · open-source · 2025-07-01

Training and Finetuning Sparse Embedding Models with Sentence Transformers

Score 1

No feed summary available yet.

Watchlist Matched: training, finetuning

Hugging Face · open-source · 2025-06-12

Post-Training Isaac GR00T N1.5 for LeRobot SO-101 Arm

Score 1

No feed summary available yet.

training agents open-source

Watchlist Matched: training, post-training

Together AI · inference-infra · 2025-05-28

Mixture-of-Agents Alignment: Harnessing the Collective Intelligence of Open-Source LLMs to Improve Post-Training

Score 3

No feed summary available yet.

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

Together AI · inference-infra · 2025-04-21

Chipmunk: Training-Free Acceleration of Diffusion Transformers with Dynamic Column-Sparse Deltas

Score 3

No feed summary available yet.

Watchlist Matched: training

Together AI · inference-infra · 2025-04-17

Together Fine-Tuning Platform, Now With Preference Optimization and Continued Training

Score 3

No feed summary available yet.

Watchlist Matched: training, fine-tuning

Hugging Face · open-source · 2025-03-26

Training and Finetuning Reranker Models with Sentence Transformers

Score 1

No feed summary available yet.

serving kernel benchmark model-release research training agents

Watchlist Matched: training, finetuning

BAIR · research · 2025-03-25

Scaling Up Reinforcement Learning for Traffic Smoothing: A 100-AV Highway Deployment

Score 6

Training Diffusion Models with Reinforcement Learning We deployed 100 reinforcement learning (RL)-controlled cars into rush-hour highway traffic to smooth congestion and reduce fuel consumption for everyone. Our goal is to tackle "stop-and...

Watchlist Matched: throughput, kernel, performance, model, paper, training, agent, agents

Replicate · inference-infra · 2024-09-20

Using synthetic training data to improve Flux finetunes

Score 0

It's easy to fine-tune Flux, but sometimes you need to do a little more work to get the best results. This post covers techniques you can use to improve your fine-tuned Flux models.