home.social

#webgpu — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #webgpu, aggregated by home.social.

  1. Another cool things is that I've learned a lot about js these last few years. I know there's a lot more to learn, but I still recall there first iteration of my library, it was super slow, but it worked! And I thought that maybe something like C# could increase speed. It was basically the same, except the #javascript was slightly faster. The real issue was that I never understood #webgl, but it got fixed with #webgpu which makes a lot more of sense to me than webgl

    #csharp #dev #development

  2. Another cool things is that I've learned a lot about js these last few years. I know there's a lot more to learn, but I still recall there first iteration of my library, it was super slow, but it worked! And I thought that maybe something like C# could increase speed. It was basically the same, except the #javascript was slightly faster. The real issue was that I never understood #webgl, but it got fixed with #webgpu which makes a lot more of sense to me than webgl

    #csharp #dev #development

  3. Another cool things is that I've learned a lot about js these last few years. I know there's a lot more to learn, but I still recall there first iteration of my library, it was super slow, but it worked! And I thought that maybe something like C# could increase speed. It was basically the same, except the #javascript was slightly faster. The real issue was that I never understood #webgl, but it got fixed with #webgpu which makes a lot more of sense to me than webgl

    #csharp #dev #development

  4. Another cool things is that I've learned a lot about js these last few years. I know there's a lot more to learn, but I still recall there first iteration of my library, it was super slow, but it worked! And I thought that maybe something like C# could increase speed. It was basically the same, except the #javascript was slightly faster. The real issue was that I never understood #webgl, but it got fixed with #webgpu which makes a lot more of sense to me than webgl

    #csharp #dev #development

  5. Another cool things is that I've learned a lot about js these last few years. I know there's a lot more to learn, but I still recall there first iteration of my library, it was super slow, but it worked! And I thought that maybe something like C# could increase speed. It was basically the same, except the #javascript was slightly faster. The real issue was that I never understood #webgl, but it got fixed with #webgpu which makes a lot more of sense to me than webgl

    #csharp #dev #development

  6. New week, more slides: Run LLMs Locally

    Now including wllama to run GGUF models inside your browser!

    wllama uses llama.cpp, WebAssembly and WebGPU, bringing a completely new experience of LLMs into the web.
    It has no 4 GB limitation and is faster than Transformers.js.

    I also added translations using the HY-MT model from Tencent.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp #webassembly

  7. New week, more slides: Run LLMs Locally

    Now including wllama to run GGUF models inside your browser!

    wllama uses llama.cpp, WebAssembly and WebGPU, bringing a completely new experience of LLMs into the web.
    It has no 4 GB limitation and is faster than Transformers.js.

    I also added translations using the HY-MT model from Tencent.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp #webassembly

  8. New week, more slides: Run LLMs Locally

    Now including wllama to run GGUF models inside your browser!

    wllama uses llama.cpp, WebAssembly and WebGPU, bringing a completely new experience of LLMs into the web.
    It has no 4 GB limitation and is faster than Transformers.js.

    I also added translations using the HY-MT model from Tencent.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp #webassembly

  9. New week, more slides: Run LLMs Locally

    Now including wllama to run GGUF models inside your browser!

    wllama uses llama.cpp, WebAssembly and WebGPU, bringing a completely new experience of LLMs into the web.
    It has no 4 GB limitation and is faster than Transformers.js.

    I also added translations using the HY-MT model from Tencent.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp #webassembly

  10. New week, more slides: Run LLMs Locally

    Now including wllama to run GGUF models inside your browser!

    wllama uses llama.cpp, WebAssembly and WebGPU, bringing a completely new experience of LLMs into the web.
    It has no 4 GB limitation and is faster than Transformers.js.

    I also added translations using the HY-MT model from Tencent.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #wllama #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp #webassembly

  11. New week, new slides: Run LLMs Locally

    Now including multi-token prediction using Qwen3.6 35B-A3B with Nextn quantization. Also speech recognition using Qwen-3-ASR is now working directly with Llama.cpp and included in the slides.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp

  12. New week, new slides: Run LLMs Locally

    Now including multi-token prediction using Qwen3.6 35B-A3B with Nextn quantization. Also speech recognition using Qwen-3-ASR is now working directly with Llama.cpp and included in the slides.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp

  13. New week, new slides: Run LLMs Locally

    Now including multi-token prediction using Qwen3.6 35B-A3B with Nextn quantization. Also speech recognition using Qwen-3-ASR is now working directly with Llama.cpp and included in the slides.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp

  14. New week, new slides: Run LLMs Locally

    Now including multi-token prediction using Qwen3.6 35B-A3B with Nextn quantization. Also speech recognition using Qwen-3-ASR is now working directly with Llama.cpp and included in the slides.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp

  15. New week, new slides: Run LLMs Locally

    Now including multi-token prediction using Qwen3.6 35B-A3B with Nextn quantization. Also speech recognition using Qwen-3-ASR is now working directly with Llama.cpp and included in the slides.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode #mtp

  16. Na jaře 2026 už WebGPU není jen záležitost Chromia nebo preview buildů. Chrome, Edge, Safari i Firefox ho dodávají v produkčních verzích, ale ne na stejných platfórmach a ne se stejnými limity. WebGPU navíc podle aktuální charty pracovní skupiny nemíří z Candidate Recommendation do W3C Recommendation. Pro vývojáře je proto důležitější konkrétní podpora, fallbacky a limity paměti než formální status standardu.

    https://zdrojak.cz/clanky/webgpu-uz-maji-vsechny-hlavni-enginy-hotovy-standard-z-nej-w3c-delat-nechce/
  17. Na jaře 2026 už WebGPU není jen záležitost Chromia nebo preview buildů. Chrome, Edge, Safari i Firefox ho dodávají v produkčních verzích, ale ne na stejných platfórmach a ne se stejnými limity. WebGPU navíc podle aktuální charty pracovní skupiny nemíří z Candidate Recommendation do W3C Recommendation. Pro vývojáře je proto důležitější konkrétní podpora, fallbacky a limity paměti než formální status standardu.

    https://zdrojak.cz/clanky/webgpu-uz-maji-vsechny-hlavni-enginy-hotovy-standard-z-nej-w3c-delat-nechce/
  18. I'll be presenting at the Cesium Developer Conference June 8–10th, on: 3D Tiles, glTF, and visualizing geospatial data on the web with Cesium.js. Consider joining! :)

    Location: Philadelphia, PA, USA
    cesium.com/events/cesium-devel
    #CesiumDevCon #WebGL #WebGPU #OpenSource

  19. I'll be presenting at the Cesium Developer Conference June 8–10th, on: 3D Tiles, glTF, and visualizing geospatial data on the web with Cesium.js. Consider joining! :)

    Location: Philadelphia, PA, USA
    cesium.com/events/cesium-devel
    #CesiumDevCon #WebGL #WebGPU #OpenSource

  20. I'll be presenting at the Cesium Developer Conference June 8–10th, on: 3D Tiles, glTF, and visualizing geospatial data on the web with Cesium.js. Consider joining! :)

    Location: Philadelphia, PA, USA
    cesium.com/events/cesium-devel

  21. I'll be presenting at the Cesium Developer Conference June 8–10th, on: 3D Tiles, glTF, and visualizing geospatial data on the web with Cesium.js. Consider joining! :)

    Location: Philadelphia, PA, USA
    cesium.com/events/cesium-devel
    #CesiumDevCon #WebGL #WebGPU #OpenSource

  22. I'll be presenting at the Cesium Developer Conference June 8–10th, on: 3D Tiles, glTF, and visualizing geospatial data on the web with Cesium.js. Consider joining! :)

    Location: Philadelphia, PA, USA
    cesium.com/events/cesium-devel
    #CesiumDevCon #WebGL #WebGPU #OpenSource

  23. New week, small update: Run LLMs Locally

    Now with a new setup for OpenCode with Qwen 3.6 and Gemma 4, including permissions and thinking variants.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

  24. New week, small update: Run LLMs Locally

    Now with a new setup for OpenCode with Qwen 3.6 and Gemma 4, including permissions and thinking variants.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

  25. New week, small update: Run LLMs Locally

    Now with a new setup for OpenCode with Qwen 3.6 and Gemma 4, including permissions and thinking variants.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

  26. New week, small update: Run LLMs Locally

    Now with a new setup for OpenCode with Qwen 3.6 and Gemma 4, including permissions and thinking variants.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

  27. New week, small update: Run LLMs Locally

    Now with a new setup for OpenCode with Qwen 3.6 and Gemma 4, including permissions and thinking variants.

    codeberg.org/thbley/talks/raw/

    #ai #llm #llamacpp #stablediffusion #qwen3 #glm #localai #gemma4 #webgpu #opencode

  28. And more than code generation, folks at Mistral have achieved some amazing results for things like speech to text, video analysis and even small conversational models running entirely on webGPU - this works even on recent iPhones! huggingface.co/mistralai/spaces

    #Mistral #webGPU

  29. And more than code generation, folks at Mistral have achieved some amazing results for things like speech to text, video analysis and even small conversational models running entirely on webGPU - this works even on recent iPhones! huggingface.co/mistralai/spaces

  30. And more than code generation, folks at Mistral have achieved some amazing results for things like speech to text, video analysis and even small conversational models running entirely on webGPU - this works even on recent iPhones! huggingface.co/mistralai/spaces

    #Mistral #webGPU

  31. And more than code generation, folks at Mistral have achieved some amazing results for things like speech to text, video analysis and even small conversational models running entirely on webGPU - this works even on recent iPhones! huggingface.co/mistralai/spaces

    #Mistral #webGPU

  32. And more than code generation, folks at Mistral have achieved some amazing results for things like speech to text, video analysis and even small conversational models running entirely on webGPU - this works even on recent iPhones! huggingface.co/mistralai/spaces

    #Mistral #webGPU

  33. This is more or less how the uniforms, storage and constants can be used now.

    You can basically assign new properties on the fly to these attributes from the main library class and that's basically it.

    #webgpu #wgsl #graphicsprogramming #javascript #webdev #webdevelopment

  34. This is more or less how the uniforms, storage and constants can be used now.

    You can basically assign new properties on the fly to these attributes from the main library class and that's basically it.

    #webgpu #wgsl #graphicsprogramming #javascript #webdev #webdevelopment

  35. This is more or less how the uniforms, storage and constants can be used now.

    You can basically assign new properties on the fly to these attributes from the main library class and that's basically it.

    #webgpu #wgsl #graphicsprogramming #javascript #webdev #webdevelopment

  36. This is more or less how the uniforms, storage and constants can be used now.

    You can basically assign new properties on the fly to these attributes from the main library class and that's basically it.

    #webgpu #wgsl #graphicsprogramming #javascript #webdev #webdevelopment

  37. This is more or less how the uniforms, storage and constants can be used now.

    You can basically assign new properties on the fly to these attributes from the main library class and that's basically it.

    #webgpu #wgsl #graphicsprogramming #javascript #webdev #webdevelopment