@xenovacom: Opus 4.7 just wrote a custom WebGPU kernel that runs Qwen3.5 up to 13x faster using a fused LinearAttention op! Agentic…
Summary
Opus 4.7 auto-generated a custom WebGPU kernel that accelerates Qwen3.5 inference up to 13× via fused LinearAttention, now shipping in Transformers.js v4.2.0.
View Cached Full Text
Cached at: 04/23/26, 02:07 PM
Opus 4.7 just wrote a custom WebGPU kernel that runs Qwen3.5 up to 13x faster using a fused LinearAttention op! Agentic kernel optimization is the future. Now live in Transformers.js v4.2.0! P.S. I’ve updated all our previous demos to use this new version. Enjoy!
Similar Articles
@ngxson: Qwen3.6-27B running 100% on WebGPU. Not the best speed but still
A developer demonstrates running the Qwen3.6-27B AI model entirely on WebGPU in a browser, though speed is not optimal.
@cniongolo: I’m not sure people realize yet that you can actually run Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated-MTP-GGUF on a dua…
Demonstrates running a custom Qwen model (Qwen3.6-35B-A3B-Claude-4.7-Opus-abliterated-MTP-GGUF) on dual Nvidia RTX PRO 6000 Blackwell GPUs at 195 tokens per second using Hugging Face Inference.
@googlegemma: “Agentic kernel optimization is the future of on-device inference” @xenovacom used Fable 5 to write kernels that pushed…
Xenova used Fable 5 to write optimized kernels achieving 255 tokens per second for Gemma 4 on WebGPU with M4, demonstrating agentic kernel optimization for on-device inference.
@TeksEdge: Solved! Qwen3.6-27B-FP8 is now running on Intel Arc Pro B70! LocalMaxxing shows a working 4× Arc Pro B70 32GB run at ~5…
Qwen3.6-27B-FP8 model is now running on Intel Arc Pro B70 GPUs at ~50 tok/s with a vLLM bug fix, marking a significant milestone for Intel GPU local AI inference.
@MiaAI_lab: Nvidia did it again! @NVIDIAAI's Qwen 3.6 27B NVFP4 is faster than Unsloth's Qwen 3.6 27B NVFP4 by a whopping ~41% on D…
Nvidia's optimized Qwen 3.6 27B NVFP4 model achieves 41% faster single-session inference and 23-25% faster concurrent inference on DGX Spark compared to Unsloth's version.