MXFP8 quantizer exceeds 6 TB/s on B200
Published on January 27, 2026, the item describes a CuTeDSL MXFP8 quantizer for B200 GPUs and reports throughput above 6 TB/s.
Read articleKnowledge you can apply
Radar
Daily bookmarks: relevant news, facts and discoveries.
Tencent presents HPC-Ops operators for LLM inference
A briefing published on January 27, 2026 describes an open-source operator library for LLM inference and reports throughput gains and faster kernels, without giving figures.
Read articleUltraRAG 3.0 defines RAG workflows in YAML
Published on January 27, 2026, UltraRAG 3.0 is described as a lightweight research framework for RAG systems, with MCP servers and configurable YAML workflows.
Read articleA series on writing a Linux debugger
Sy Brand published a series about writing a debugger for Linux, collected under the “Debuggers” tag on the author’s blog. The topic may interest engineers working on debugging tools.
Read articleINTELLECT-2 points to delta=4.0 for steadier GRPO
Published on January 26, 2026, the item says the INTELLECT-2 technical report introduced a setting intended to improve GRPO training stability: use delta=4.0. It is a concrete parameter engineers can investigate.
Read articleRadar: selected resources on agentic reasoning
A GitHub list gathers papers and resources based on a survey about agentic reasoning for language models, aimed at engineers researching the topic.
Read articleCornell introductory compiler notes
Cornell offers introductory compiler notes for its CS 4120/ENGRD 4120 course, taught in spring 2022.
Read articleCALM proposes predicting vector blocks
Published on January 24, 2026, the item describes CALM models that predict vectors for meaning blocks rather than individual tokens. It claims 4× more information per step and a 44% reduction in training compute.
Read articleCLI estimates VRAM needed to load models
Published January 23, 2026, this item describes a command-line tool that estimates the VRAM needed to load a model without loading it.
Read articleUnsloth guide to embedding fine-tuning
Published on January 22, 2026, the material explains embedding model fine-tuning and includes a Colab notebook for EmbeddingGemma (300M).
Read articleQwen3-TTS announced as open source
On January 22, 2026, a post announced that Qwen3-TTS would be made available as open source and mentioned voice cloning and design. It provided no repository or other link.
Read articleFine-tuning a 14B code model with SFT and DPO
Published on January 20, 2026, the article reports fine-tuning a 14B model on code conversations with SFT and DPO, and discusses results, costs, and difficulties.
Read articleslime v0.2.2 adds Int4-QAT and R3 support
Published on January 18, 2026, the slime v0.2.2 entry reports memory and performance improvements, Int4-QAT training, and full R3 support with DeepEP and MTP.
Read articleArchitecture influences retention and forgetting
A January 17, 2026 note on a comparison of architectures for continual learning: wider networks and less global average pooling may favor retention, while ResNets and WideResNets learn new tasks quickly but forget more.
Read articleStudy examines attacks on Federated Learning
A study of gradient inversion attacks reports that the most practical attack method remains unreliable and proposes a three-stage defense.
Read articleSAR as a reward for LLM RL
A January 16, 2026 publication presents Self-Aligned Reward (SAR) as a reward signal compatible with PPO and GRPO, intended to improve reasoning and reduce tokens.
Read articleStep Audio R1.1: real-time spoken dialogue
A January 16, 2026 publication presents Step Audio R1.1 as an interactive real-time spoken-dialogue model and describes its license and architecture.
Read articleUnsloth describes GRPO with 380K context
Published on January 15, 2026, the item summarizes documentation on long-context reinforcement learning fine-tuning with GRPO. Unsloth says its batching algorithms enable 380K context for gpt-oss on a 192 GB GPU.
Read articleRadar: controlled evolution for code optimization
A January 14, 2026 article proposes an iterative method to improve algorithmic code using diverse strategies, feedback-guided mutations, and experience across tasks. The approach is evaluated on EffiBench-X.
Read articleBayesian generalization in offline RL
A January 14, 2026 post presents a paper on adaptive generalization in model-based offline reinforcement learning and long-horizon rollouts.
Read articleRota Nacional
Bring privacy into your workflow.
30 days, no card, with a starting quota. After that, Pix credit from R$ 10,00.