b9840

parruda/swarmb9840Jun 29, 2026by github-actions[bot]
AI Summary

Adds full support for DeepSeek V4 models, featuring conversion tools, graph-based processing, multi-sequence support, and state save/load capabilities.
Key Highlights

Added DeepSeek V4 model support (including Pro model)
Implemented graph-based processing and FA (Flash Attention) support
Added save and load state functionality
Added multi-sequence support
New Features

DeepSeek V4 model support (conversion, graph input, chat templates)
State save/load
Multi-sequence support
Full Release Notes

<details open>

DeepSeek V4  (#24162)

* convert: add dsv4 conversion

* add basic setup

* add llm_graph_input_dsv4

* add save-load state

* add sinkhorn eps - correction by @fairydreaming

* add rope fix

* cleanup dead code

* fix bugs

* support pro model: added by @fairydreaming

* remove redundant V cache

* Chat template

* remove debugging leftovers

* Add mechanism for inlining templates based on architecture

* s/deepseek-v4-flash/deepseek4/g

* s/deepseek-v4-flash/deepseek4/g continued

* enable graph reuse

* enable FA

* fix test llama archs

* rename

* compatibility with antirez ds4 GGUFs

* simplified set_gguf_parameters() by calling super class method, replaced moe.score_func with expert_gating_func.

* reserve worst-case kv-cache

* revert max split inputs

* address review comments

* add padding to enable FA

* pad only the final value of plan.n_kv to 256

* remove built-in cpp chat template

* cont: remove cpp built-in template

* rm outdated test

* replace ggml_view_3d() with ggml_reshape_3d()

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* only support n_seq=1 for now

* remove unused var

* cont: remove unused var

* use scale bias

* use correct ptr for can_reuse

* remove gen-chat-inline-templates.py

* simplify graph reuse

* cont: cleanup

* remove unused inputs

* enable partial checkpointing

* add correct shape for kq_mask + set llama_model_n_swa to 0 for dsv4

* precompute source_idx + add comment about dummy write

* support multi-seq

* remove restored_trim_pos

* use split_equal when possible

* fix indent

* address review comments

* use LLM_KV

* fix ci

---------

Co-authored-by: Piotr Wilkin <piotr.wilkin@syndatis.com>
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: fairydreaming <166155368+fairydreaming@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

</details>

**macOS/iOS:**
- [macOS Apple Silicon (arm64)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-macos-arm64.tar.gz)
- macOS Apple Silicon (arm64, KleidiAI enabled) [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23780)
- [macOS Intel (x64)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-macos-x64.tar.gz)
- [iOS XCFramework](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-xcframework.zip)

**Linux:**
- [Ubuntu x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-x64.tar.gz)
- [Ubuntu arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-arm64.tar.gz)
- [Ubuntu s390x (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-s390x.tar.gz)
- [Ubuntu x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-vulkan-x64.tar.gz)
- [Ubuntu arm64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-vulkan-arm64.tar.gz)
- [Ubuntu x64 (ROCm 7.2)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-rocm-7.2-x64.tar.gz)
- [Ubuntu x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-openvino-2026.2.1-x64.tar.gz)
- [Ubuntu x64 (SYCL FP32)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-sycl-fp32-x64.tar.gz)
- [Ubuntu x64 (SYCL FP16)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-ubuntu-sycl-fp16-x64.tar.gz)

**Android:**
- [Android arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-android-arm64.tar.gz)

**Windows:**
- [Windows x64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-cpu-x64.zip)
- [Windows arm64 (CPU)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-cpu-arm64.zip)
- [Windows arm64 (OpenCL Adreno)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-opencl-adreno-arm64.zip)
- [Windows x64 (CUDA 12)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-cuda-12.4-x64.zip) - [CUDA 12.4 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9840/cudart-llama-bin-win-cuda-12.4-x64.zip)
- [Windows x64 (CUDA 13)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-cuda-13.3-x64.zip) - [CUDA 13.3 DLLs](https://github.com/ggml-org/llama.cpp/releases/download/b9840/cudart-llama-bin-win-cuda-13.3-x64.zip)
- [Windows x64 (Vulkan)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-vulkan-x64.zip)
- [Windows x64 (OpenVINO)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-openvino-2026.2.1-x64.zip)
- [Windows x64 (SYCL)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-sycl-x64.zip)
- [Windows x64 (HIP)](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-bin-win-hip-radeon-x64.zip)

**openEuler:**
- [DISABLED](https://github.com/ggml-org/llama.cpp/pull/23705)
- openEuler x86 (310p)
- openEuler x86 (910b, ACL Graph)
- openEuler aarch64 (310p)
- openEuler aarch64 (910b, ACL Graph)

**UI:**
- [UI](https://github.com/ggml-org/llama.cpp/releases/download/b9840/llama-b9840-ui.tar.gz)