Skip to content
AI NEWS JOURNAL
AI, IN PERSPECTIVE

Search AI news

Search headlines, summaries and articles to connect the dots.

Newest first

Amazon Polly launches voice cloning: A unique voice for your brand

Voice cloning in Amazon Polly is now generally available. Give Amazon Polly a few minutes of recordings from a consenting speaker, and it builds a custom voice that reads new text in that speaker's voice. Cloned voices run on the Amazon Polly generative engine and support seven languages.

Amazon SageMaker Inference: 2026 launches in review, part 2

This second installment reviews 13 more Amazon SageMaker AI inference launches from 2026, spanning configuration recommendations, fitting more models onto each GPU, serving models too large for one instance, cross-Region routing, and post-launch tuning, bringing the year's total to 26.

b12543

The update introduces a new tensor API conv_2d kernel for Metal, enabling split-K optimizations across all chips, including M5. It improves efficiency by utilizing tensor cores, reducing idle GPU time during long reductions. The kernel now splits reductions into partial sums stored in buffers, with a follow-up kernel summing them deterministically. Address math is optimized using 32-bit byte offsets, reducing computational overhead. A fallback to the generic kernel is implemented for images exceeding 4 GiB. Performance benchmarks show significant speed improvements on M5 Max devices, with reduced execution times across various convolution configurations.

b12549

The llama.cpp v1.25.49 release includes fixes for Inkling 2 attention mask handling, making the relative-position input a hybrid iswa input, enabling optional expert weight scale and final logit softcap, and improving KV rows mask generation for hybrid iswa memory. The update also standardizes code formatting in inkling2.cpp and ensures compatibility across multiple platforms including macOS, Linux, Windows, Android, and openEuler.

b12552

The update includes fixes for OpenVINO CI, optimizations for model recompiles in the prefill stage, improvements to output handling during model reuse, and support for f16 SCALE and f16/bf16 TRI operations. It also addresses issues with static-shape views in MoE layers, enhances SSM_SCAN build checks, and adds Falcon-H1 to the list of validated OpenVINO models with noted limitations.

b12556

A vulnerability in llama.cpp's HRM layer count validation allows integer overflow due to improper handling of uint32_t multiplication, leading to potential buffer overflows during tensor loading. The issue arises when the product of n_hrm_layers_per_stack, n_hrm_h_cycles, and (n_hrm_l_cycles + 1) is computed as a uint32_t and compared to block_count, enabling maliciously crafted inputs to pass validation despite exceeding the actual block count.

b12558

A fix has been implemented to disable subgroup arithmetic in the Vulkan driver for Qualcomm's Adreno GPUs, addressing an issue where subgroupAdd() after divergent code returns incorrect sums. This correction resolves problems with the decode-step norm, preventing models from generating stray tokens after long prompts. Testing on Adreno 850 and X2-90 showed improved results for ADD_RMS_NORM and RMS_NORM_MUL_ADD operations, while performance on llama 3B Q4_0 remained unchanged.

b12561

The update introduces a chunked GATED_DELTA_NET implementation for Vulkan, including shader additions, optimizations for RDNA4 and Ampere architectures, fixes for shared memory bank conflicts and NVIDIA shader compiler issues, and support for various subgroup sizes and backend configurations. It also enables the feature by default and adds diagnostic markers for device information.

b12563

The update addresses a precision issue in CUDA's per-expert fallback path where the MUL_MAT_ID operation was not properly preserving precision due to an unset operation type. The fix sets the operation to GGML_OP_MUL_MAT and reverts an earlier change that forced fp32 output, ensuring consistent precision handling across quantized matrix multiplication paths. This resolves issues with overflow to infinity in fp16 when results exceed 65504, particularly affecting models using IQ1_M weights with large batches.

b12567

The update adds QK data type support for s390x architecture in llamafile, along with cleanup improvements for s390x compatibility.

Introducing enhanced infrastructure observability in Amazon SageMaker HyperPod

Amazon SageMaker HyperPod now offers enhanced observability to help you diagnose GPU training issues faster. It brings together diagnostic and profiling capabilities for training workloads on both Amazon EKS and Slurm orchestrators, pinpointing issues across GPU, network, and host layers.

Optimize AI economics on Amazon Bedrock with Accenture’s sequenced approach

As enterprises scale AI on Amazon Bedrock, token spend can grow even as per-token prices fall. This post walks through six composable optimization levers, shows how Accenture applied them across four client engagements, and recommends a sequence for cutting AI token spend on your own workloads.

PolyAI Dialog-RSN-1 is now on Amazon SageMaker JumpStart

PolyAI Dialog-RSN-1, a reasoning model built for real-time complex dialog, is available today on AWS Marketplace and Amazon SageMaker JumpStart. Discover it, evaluate it on your own data, and deploy it to a secure SageMaker endpoint in your own AWS account.

Protect agent tool calls with Skyflow and Amazon Bedrock AgentCore Gateway

Learn how Skyflow and Amazon Bedrock AgentCore Gateway keep sensitive data out of the tools your AI agents call. Added as a gateway interceptor, Skyflow de-identifies data before any tool, model, or log sees it, and restores it for authorized callers, with no change to your agent code.

Use Kimi Code CLI with models on Amazon Bedrock

Kimi Code CLI v2.7.0 adds a native Amazon Bedrock provider, so you can connect the open source coding agent to models like Kimi K3.1 Pro using your standard AWS credential chain (named profiles or SSO) instead of a separate API key. This post shows how to configure the provider and send your first request.