Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

v1.0.0

The llama.cpp project has released version 1.0.0, marking a significant milestone with numerous improvements across multiple platforms including CUDA, Vulkan, Metal, SYCL, OpenCL, and WebGPU. Key updates include enhanced support for quantized models (Q5_0, Q5_1, FP8), optimizations for flash attention and MoE caching, improved server security features like CORS protection and SSRF mitigation, UI enhancements such as sidebar reorganization and live-preview markdown editing, and better cross-platform compatibility. The release also includes bug fixes, CI/CD improvements, and updated dependencies.

Inkling 2 Small on vLLM: Day-0 Support

vLLM supports Thinking Machines' Inkling 2 Small from day 0: attention residuals, a new decode kernel for relative-bias attention, all-reduce fused with the ops around it, and DSpark speculative decoding.

Integrating SGLang into Spyre in Under a Week with AI

TL;DR With help from AI coding agents, our two-person team integrated SGLang with IBM Spyre in less than a week, building a plugin of roughly 1,500 lines. Existing Spyre support... The post Integrating SGLang into Spyre in Under a Week with AI appeared first on PyTorch.

The Definitive Fine-Tuning Guide for 3D Generation Models

This blog post provides scripts and Docker containers for fine-tuning TRELLIS.2, a 3D generation model, using LoRA adapters on footwear data from the Google Scanned Objects (GSO) dataset. The approach focuses on improving geometric accuracy by training only the sparse-structure and shape flow diffusion transformers while freezing the texture stage. The process includes preprocessing with Blender to generate 16 views per object, converting meshes into latents, and training with Hugging Face Jobs or local Docker. Results show significant improvements in F-score, chamfer distance, and normal consistency on held-out test objects, demonstrating effective adaptation for specific 3D object categories.

Continuous Angular Control in Text-to-Image Generation via Geometric Token Encoding

arXiv:2611.13172v1 Announce Type: new Abstract: Text-to-image diffusion models can render a photorealistic analog clock but cannot control the time it displays. Across four pretrained models, fewer than 1% of generated clocks place both hands within 6 degrees of the requested time; the models instead collapse onto visually dominant configurations such as the canonical 10:10 pose. The same failure appears when a specific fraction of a circular object is requested, where models return complete or conventionally sliced objects. We observe that these two failures share a geometric structure: both target quantities can be parameterized by angular variables. We introduce a geometric token encoding framework that conditions generation directly on these angles. Each angular component is encoded as a periodic sine-cosine feature and projected into a role-specific token embedding that is injected into the prompt before a frozen text encoder. To spatially ground these tokens without requiring localization input at inference, we introduce Stochastic Localization Relaxation (SLR), which stochastically relaxes attention supervision from component-level to object-level to unconstrained localization during training. For clocks, we additionally introduce a hand-swap preference objective that discourages assigning the correct angles to the wrong hand roles. On analog clocks our method reaches 63-78% Acc@6 degrees across evaluation methods, compared with below 1% for the baselines. On circular-fraction generation, it achieves 89.5% Acc@0.10, while maintaining 88.8% on unseen object categories and 84.9% on unseen target values, compared with roughly 20% for pretrained baselines.

Optimizing Muse Spark Open on TPU v7 with SGLang-TorchTPU

TL;DR We serve Meta’s Muse Spark Open, an 816B-parameter mixture-of-experts model, on a single TPU v7 host of four chips, with SGLang running on the TorchTPU PyTorch backend. At the... The post Optimizing Muse Spark Open on TPU v7 with SGLang-TorchTPU appeared first on PyTorch.

v0.45.0

ComfyUI v0.45.0 introduces multiple updates including new partner nodes for Pika, Tencent, Kling, OpenAI, Qwen, Gemini, Grok, Breeze, and ByteDance; performance enhancements for Qwen 3.x; bug fixes for crashes, database issues, and VAE selection errors; improved memory management with adaptive slice lengths; support for new models like Seedance 2.5, Seedream 5.5, and MiniMax-H3-image; added video input for Qwen 3.x, solid mode in mesh extraction, and texture rebaking; updated workflow templates to v0.11.105 and API documentation; default activation of local_web_sso and partner_node_token; and contributions from new developer GACLove.

b12246

On Qualcomm Adreno GPUs, scalar flash attention was limited to 8 rows per workgroup when head size exceeds 128 to prevent register spilling and performance degradation. This change improved performance significantly, reducing execution time from several seconds to milliseconds and enabling completion of the test-backend-ops case.

Indie Game Development with AI

Article URL: https://www.sandromaglione.com/articles/indie-game-development-with-ai Comments URL: https://news.ycombinator.com/item?id=50432523 Points: 22 # Comments: 16

Meet the Sad Wives of AI

Article URL: https://www.wired.com/story/meet-the-sad-wives-of-ai/ Comments URL: https://news.ycombinator.com/item?id=50438429 Points: 27 # Comments: 13

NVIDIA Nemotron 3.5 Super VL is now available in Amazon SageMaker JumpStart

NVIDIA Nemotron 3.5 Super VL, a 120B-parameter multimodal vision-language model, is now available in Amazon SageMaker JumpStart. Deploy the NVFP4 or BF16 variant in a few steps, then run text, image, and video inference for document intelligence, computer-use agents, and multimodal agentic workloads.

OneReel: Efficient Multi-Scene Video Generation with Scene-Structured Attention

arXiv:2611.10594v1 Announce Type: new Abstract: Multi-scene video generation must follow scene-specific prompts while preserving consistency across scenes. Dense joint generation enables cross-scene information exchange but incurs quadratic attention cost, whereas separate generation avoids this cost at the expense of cross-scene consistency. We present OneReel, a training-free system that exploits scene structure to enable efficient multi-scene generation. Our key insight is that multi-scene generation can preserve complete intra-scene attention while restricting cross-scene information exchange to a bounded shared memory. OneReel uses scene prompts and durations to establish scene-specific attention scopes and dynamically selects a bounded set of representative latent frames for cross-scene sharing. This reduces video attention complexity from quadratic to linear in the number of scenes. To translate these savings into execution speedups, OneReel packs scene-local and shared K/V into dense variable-length attention and employs work-balanced multi-GPU partitioning with communication-aware selection between K/V gathering and Ulysses-style parallelism. Our evaluations across four video generation models show that OneReel achieves 1.7-4.9x end-to-end speedup over dense attention and 1.2-2.2x over existing sparse-attention advances, while improving text-following accuracy by 34-37 percentage points and maintaining cross-scene consistency.