#opencl
TornadoVM 8 dropped Graal and JVMCI. Our new compiler, TC, is written in plain #Java.

Our new video shows TC’s five stages, from Java Bytecode to #CUDA, #OpenCL, #Metal (and soon #HIP). Compiling needs no #GPU at all.

www.youtube.com/watch?v=W0oH...

#opensource #TornadoVM #compiler
TC: TornadoVM’s Own Compiler, from Java Bytecode to CUDA, OpenCL and Metal
YouTube video by TornadoVM
www.youtube.com
November 20, 2026 at 12:04 PM
Run lattice Boltzmann CFD on any GPU or CPU via OpenCL

Explore it here:
osp.fyi/fluidx3d
November 17, 2026 at 8:30 AM
cuda was relevant Back Then. doesnt really matter nowadays in my experience. everything supports HIP/SyCL/OpenCL or even straight Vulkan
November 16, 2026 at 9:03 AM
Daily Bitcoin card · 20 Nov 2026

BTC closed yesterday at $61,914 (▼ 1.7% 24h, ▼ 25.4% 30 days).

bitcoin.onl/today
November 20, 2026 at 1:01 PM
Offline on-device LLM chat on Google Play required CPU-first execution, GPU fallback handling, Android 12 OpenCL manifest fixes, and content-safe chat template parsing.
Save What Matters
Curate Feeds | Make Collections | Customize Email Briefs
briefly.co
November 20, 2026 at 1:26 AM
Offline on-device LLM chat on Google Play required CPU-first execution, GPU fallback handling, Android 12 OpenCL manifest fixes, and content-safe chat template parsing.
Save What Matters
Curate Feeds | Make Collections | Customize Email Briefs
briefly.co
November 20, 2026 at 1:25 AM
Il y a des outils qui utilisent les outils de Nvidia, qui a poussé tout le monde à utiliser son logiciel propriétaire, Cuda.
Une fois que vous êtes sorti de ce piège (par exemple parce que Nvidia ne vous finance plus), vous pouvez utiliser n'importe quelle architecte, généralement avec OpenCL.
November 21, 2026 at 10:01 PM
This day in Bitcoin history · November 20: Re: OpenCL miner for the masses (2010)

Quote from: m0mchil on November 20, 2010, 10:16:19 AM updated to SVN 186 Thanks m0mchil for keeping up on the updates! GPU miners, please upgrade as soon as possible to shut down the free transaction abuse! This…
This day in Bitcoin history · November 20: Re: OpenCL miner for the masses (2010)
Quote from: m0mchil on November 20, 2010, 10:16:19 AM updated to SVN 186 Thanks m0mchil for keeping up on the updates! GPU miners, please upgrade as soon as possible to shut down the free transaction abuse! This versi…
bitcoin.onl
November 20, 2026 at 1:01 PM
Mamy Ratsimbazafy will show what that unlocks in Tattletale, the AI inference engine they are building: GPU kernels for CUDA, OpenCL, Vulkan and WebGPU generated at compile time, shipped as a single binary with no dependency hell, and usable from Python, embedded in C, or compiled to JavaScript/WASM
November 20, 2026 at 3:30 PM
he could do the funniest thing and spill beans about how everything Nvidia is doing boils down to worldwide market manipulation driven by hype for an ill-fated technology monopolized by their hardware because CUDA wasn't an open standard like OpenCL so AMD could never compete with them
“.. After weeks spent denying the allegations of infidelity and fighting to remain in his job .., he’d started using phrases like ‘whatever’ and ‘I don’t care.’ .. castigating himself as a ‘piece of sh--.’”

@bloomberg.com #$NVDA
www.bloomberg.com/news/article...
November 13, 2026 at 8:48 AM
llama.cpp changes target local inference, not new model weights.

Initial OpenCL support for iq3_s/iq3_xxs includes Adreno paths; fixes address Hexagon DMA strides ≥16 MB and InternVL CLS-token order. Backend support alone isn’t production readiness: test quality, latency…

Read more on Kimbodo:
What New llama.cpp Updates Mean for Reliable Local AI Inference
What Happened The substantiated developments are in llama.cpp tooling, not new open-weight model releases. Recent changes add initial OpenCL support for the iq3_s and iq3_xxs quantization formats, including Adreno-specific execution…
kimbodo.com
November 17, 2026 at 6:16 PM
this fun new photoshop bug is the bane of my existence. am i being paranoid or is adobe shipping more sloppppppp community.adobe.com/bug-reports-...
P: PS 28.0 Align buttons in Move tool options bar stay disabled after Cmd+click layer selection + Select All | Community
Environment- Adobe Photoshop 2027, version 28.0.0 (20261027.r.20 3951c45), arm64- macOS 27.2.0- Apple M4 Max, 64 GB RAM- GPU: Apple M4 Max (GPU and OpenCL enabled)- Display: Studio Display 5K, HDR ena...
community.adobe.com
November 17, 2026 at 8:55 PM
jitLLM 3.0 is out! #Java LLM inference on CUDA, Metal and OpenCL #GPUs via TornadoVM.

• 1.43× faster startup with @graalvm.org Native Image
• Up to 9.3× faster prefill
• Zero-copy weights on Apple silicon

Soon in @langchain4j.dev, @quarkus.io & @spring.io #SpringAI.

github.com/beehive-lab/jitllm
November 13, 2026 at 11:06 AM
CoffeeLoader Malware Returns With a New GPU-Powered Armoury Packer Variant + Video

A redesigned infection mechanism replaces OpenCL with OpenGL, strengthens persistence, and uses legitimate Windows processes to conceal malicious activity. Introduction: A Malware Loader Evolves to Evade Detection…
CoffeeLoader Malware Returns With a New GPU-Powered Armoury Packer Variant + Video
A redesigned infection mechanism replaces OpenCL with OpenGL, strengthens persistence, and uses legitimate Windows processes to conceal malicious activity. Introduction: A Malware Loader Evolves to Evade Detection Cybersecurity researchers have uncovered a new variant of CoffeeLoader, a sophisticated malware loader designed to deliver additional malicious payloads onto compromised Windows systems. The latest discovery, documented by Kaspersky, highlights significant changes to the malware's Armoury packer, including a redesigned GPU-assisted decryption mechanism, a new shellcode-based delivery process, and stronger persistence capabilities.
undercodenews.com
November 16, 2026 at 2:23 PM
The beta of OpenMM 8.7 is now available. New features include a Metal platform (for Apple GPUs), a tool for packing boxes of arbitrary liquid mixtures, and anisotropic barostats for RPMD simulations. Please test it out and let us know if you find any issues: github.com/openmm/openm...
OpenMM 8.7 beta · openmm openmm · Discussion #5603
The beta of OpenMM 8.7 is now online. You can install it with mamba install -c conda-forge/label/openmm_rc -c conda-forge openmm or pip install --pre openmm The base pip package includes the OpenCL...
github.com
November 18, 2026 at 7:52 PM
Khronos Compute is now open. It is a new Discord community for developers working with SYCL and OpenCL, where you can ask questions, share projects and talk directly with working group members. Join us and tell us what you are building: khr.io/computediscord.
#OpenCL #SYCL #Compute #support
Join the Khronos Compute Discord Server!
Check out the Khronos Compute community on Discord - hang out with 37 other members and enjoy free voice and text chat.
khr.io
November 12, 2026 at 4:52 PM
Russian drones using Nvidia's chips, and Nvidia's proprietary language (Cuda)? This would be grossly stupid. OpenCL is a portable equivalent in the PC world, and inexpensive microcontrollers with NPUs might be sufficient for a drone.
November 14, 2026 at 11:53 AM
Khronos Compute is now open. It is a new Discord community for developers working with SYCL and OpenCL, where you can ask questions, share projects and talk directly with working group members. Join us and tell us what you are building: khr.io/computediscord.
#OpenCL #SYCL #Compute #support
Join the Khronos Compute Discord Server!
Check out the Khronos Compute community on Discord - hang out with 37 other members and enjoy free voice and text chat.
khr.io
November 12, 2026 at 4:53 PM
Khronos Compute is now open. It is a new Discord community for developers working with SYCL and OpenCL, where you can ask questions, share projects and talk directly with working group members. Join us and tell us what you are building: https://khr.io/computediscord.
#opencl #sycl #compute #support
November 12, 2026 at 4:53 PM
📰 **Khronos To Explore New High Level API For Remote Offloading**

The Khronos Group, the industry consortium responsible for OpenGL, OpenCL, Vulkan, and other industry standard APIs has formed an exploratory group to evaluate the possibility of a high-level remot...

📰 Source: Phoronix
🔗 Link […]
Original post on igeek.gamer-geek-news.com
igeek.gamer-geek-news.com
November 14, 2026 at 3:43 PM
AMD has been trying real hard with Vulkan and OpenCL, but it still has a lot to catch up with.

If the Chinese companies also go Vulkan and OpenCL, Nvidia loses its moat. Then it doesn't make a hot item it makes a commodity. And wall street hates commodities. Which means he ain't rich then
November 12, 2026 at 4:32 PM