abetlen
diff --git a/CHANGELOG.md b/CHANGELOG.md
@@ -8,7 +8,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ## [Unreleased]
 
 - feat(example): support server video inputs and Gemma text tool calls by @abetlen in #2291
-- feat: update llama.cpp to ggml-org/llama.cpp@8f83d6c27
+- feat: update llama.cpp to ggml-org/llama.cpp@e3471b3e7
 - fix(example): support multi-step Responses tool streaming by @abetlen in #2288
 - fix(ci): Repair Linux accelerator wheels for manylinux publishing
 

diff --git a/vendor/llama.cpp b/vendor/llama.cpp
+1 −1		.devops/cpu.Dockerfile
+1 −1		.devops/cuda.Dockerfile
+1 −1		.devops/intel.Dockerfile
+1 −1		.devops/musa.Dockerfile
+1 −1		.devops/openvino.Dockerfile
+1 −1		.devops/rocm.Dockerfile
+1 −1		.devops/vulkan.Dockerfile
+1 −1		.devops/zendnn.Dockerfile
+23 −0		.github/workflows/build-webgpu.yml
+10 −0		conversion/gemma.py
+0 −1		ggml/src/ggml-cuda/mmvq.cu
+25 −11		ggml/src/ggml-webgpu/ggml-webgpu-shader-lib.hpp
+69 −44		ggml/src/ggml-webgpu/ggml-webgpu.cpp
+8 −5		ggml/src/ggml-webgpu/wgsl-shaders/binary.wgsl
+19 −1		ggml/src/ggml-webgpu/wgsl-shaders/concat.wgsl
+267 −543		ggml/src/ggml-webgpu/wgsl-shaders/mul_mat_decls.tmpl
+6 −4		ggml/src/ggml-webgpu/wgsl-shaders/scale.wgsl
+7 −4		ggml/src/ggml-webgpu/wgsl-shaders/unary.wgsl
+6 −0		gguf-py/gguf/constants.py
+8 −0		gguf-py/gguf/tensor_mapping.py
+4 −0		src/llama-arch.cpp
+3 −0		src/llama-arch.h
+13 −4		src/llama-graph.cpp
+3 −0		src/models/gemma4-assistant.cpp
+3 −0		tools/server/server-task.cpp