* Prefer vector flash decoding kernel for Gemma models
Vector flash decoding kernel was not being picked for models with head dimension 256. Gemma models are in this category.
Removing this limit improves e2e performance by upto 12% in gen phase throughput for Gemm models.
* Update ggml/src/ggml-cuda/fattn.cu
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* (wip) refactor downloading system [no ci]
* fix all examples
* fix mmproj with -hf
* gemma3: update readme
* only handle mmproj in llava example
* fix multi-shard download
* windows: fix problem with std::min and std::max
* fix 2
This patch enables usage of MMA when one of the
dimensions of the matrix(ie either M or N) is 1. This
is useful in case of token generation where N < 2.
The concept of 'GEMV Forwarding' is used where when one
of the matrix has a single row/column, the elements are
broadcasted, instead of using packing routine to prepack
the matrix elements.
This change results in 5% - 15% improvement in total
speed(ie all tokens/total time), across various batch
sizes. This is in comparision with the corresponding
dot product implementation.
The patch is tested with FP32 models of Meta-Lllama-3-8B,
Mistral-7B, Llama-2-7B-chat-hf on a IBM POWER10 machine.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com>
This change upstreams llamafile's cpu matrix
multiplication kernels for ppc64le ISA using MMA
builtins. This patch handles matrix multiplication
between quantised datatypes, block_q4_0 and
block_q8_0.
This change results in 5% - 50% improvement
in total speed(ie all tokens/total time), across
various batch sizes.
The patch is tested with Meta-Lllama-3-8B,
Mistral-7B, Llama-2-7B-chat-hf models on a
IBM POWER10 machine.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com>
* webui: Make textarea uncontrolled to eliminate devastating lag
* Update index.html.gz
* use signal-style implementation
* rm console log
* no duplicated savedInitValue set
---------
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
* Add block interleaving support for Q4_K quantization
* Remove whitespaces and fix CI/CD issues
* Update pointer of bsums from int16_t to const int16_t
* Add vector version of quantize_q8_K_4x8 function
* Update code formatting based on review comments
- Find out active blocks per SM using cudaOccupancyMaxActiveBlocksPerMultiprocessor API. Use this value to determine the optimal parallel_blocks value.
- Prefer vector flash attention kernels over MMA kernel for BS=1
Fixes Issue: #12182
---------
Co-authored-by: Johannes Gäßler <johannesg@5d6.de>
* Enable CUDA Graph on CTK < 12.x
`cudaGraphExecUpdate` API was changed on 12.x. For this reason CUDA graph support was disabled on older CUDA toolkit. This change enables CUDA support in CTK version < 12.x by using older API if CTK < 12.x.
* Fix compilation errors with MUSA
* Disable CUDA Graph for MUSA
* metal : refactor im2col parameters into a struct
* metal: Change im2col offset types from int32_t to uint64_t to support larger memory offsets
* metal : refactor sum_rows parameters into a struct
* metal : refactor soft_max parameters into a struct
* metal : refactor diag_mask_inf parameters into a struct
* metal : refactor ssm_conv parameters into a struct
* metal : refactor ssm_scan parameters into a struct
* metal : refactor get_rows parameters into a struct
* metal : refactor group_norm parameters into a struct
* metal : refactor conv_transpose_1d parameters into a struct
* metal : refactor upscale parameters into a struct
* metal : refactor pad parameters into a struct
* metal : refactor pad_reflect_1d parameters into a struct
* metal : refactor arange parameters into a struct
* metal : refactor timestep_embedding parameters into a struct
* metal : refactor argsort parameters into a struct
* metal : refactor leaky_relu parameters into a struct
* metal : refactor pool_2d parameters into a struct
* metal : fix trailing whitespace
---------
Co-authored-by: alexju <alexju@tencent.com>
* Added Phi-4-mini-instruct support
* Update regex per ngxson
* Change the vocab base to Xenova/gpt-4o
* fix conversion update script
* no need to check longrope
* minor style fix
* fix python style
---------
Co-authored-by: Nicholas Sparks <nisparks@microsoft.com>
* server : add TEI API format for /rerank endpoint
* Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* fix
* also gitignore examples/server/*.gz.hpp
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Bug fix for clamp_f32
When using tensors larger than 1d clamp operation does not work due to the restriction of returning if ith is not 0.
* Bug fix for clamp_f32
* Bug fix for clamp_f32
* server : (webui) introduce conversation branching + idb storage
* mark old conv as "migrated" instead deleting them
* improve migration
* add more comments
* more clarification
* redo Settings modal UI
* add python code interpreter
* fix auto scroll
* build
* fix overflow for long output lines
* bring back sticky copy button
* adapt layout on mobile view
* fix multiple lines output and color scheme
* handle python exception
* better state management
* add webworker
* add headers
* format code
* speed up by loading pyodide on page load
* (small tweak) add small animation to make it feels like claude
* init version
* fix auto scroll
* bring back copy btn
* bring back thought process
* add lint and format check on CI
* remove lang from html tag
* allow multiple generations at the same time
* lint and format combined
* fix unused var
* improve MarkdownDisplay
* fix more latex
* fix code block cannot be selected while generating
* Added quantization for visual projector
* Added README
* Fixed the clip quantize implementation in the file
* Fixed the gcc warning regarding minor linting
* Removed trailing whitespace
* Add SVE support for q4_K_q8_K
* Update ggml/src/ggml-cpu/ggml-cpu-quants.c
change to use K_SCALE_SIZE
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
This change upstreams llamafile's cpu matrix
multiplication kernels for ppc64le using MMA
builtins for quantised int8 datatype.
This change results in 10% - 70% improvement
in total speed(ie all tokens/total time), across
various batch sizes.
The patch is tested with Meta-Lllama-3-8B,
Mistral-7B, Llama-2-7B-chat-hf models on a
IBM POWER10 machine.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com>
* Update cmakepreset.json to use clang with ninja by default
* Update cmakepreset.json to add clang and ninja based configs
* Updates to build.md file
* Make updates to rename preset targets
* Update with .cmake file
* Remove additional whitespaces
* Add .cmake file for x64-windows-llvm
* Update docs/build.md
* Update docs/build.md
---------
Co-authored-by: Max Krasnyansky <max.krasnyansky@gmail.com>
* CANN: Fix the bug build fail on Ascend310P under two cases:
1) Manual specify SOC_TYPE
2) Under some unusual compile environment
* Update the cann backend News content: Support F16 and F32 data type model for Ascend 310P NPU.
* fix CANN compile fail bug: the assert in ascend kernel function doesn't supportted on some CANN version
* CANN Support Ascend310P to accelerate F32 and F16 Model
* Add compile option soc type macro ASCEND_310P to ggml-cann lib
* Remove unused code
* Remove the ascend soc_type hard code compile option in CMakelist.txt
This change upstreams llamafile's cpu matrix
multiplication kernels for ppc64le using MMA
builtins for FP32 datatype.
This change results in a consistent 90%
improvement in input processing time, and 20%
to 80% improvement in output processing time,
across various batch sizes.
The patch is tested with Meta-Lllama-3-8B,
Mistral-7B, Llama-2-7B-chat-hf models on a
IBM POWER10 machine.
Signed-off-by: Amrita H S <amritahs@linux.vnet.ibm.com>
* Add granite template to llama.cpp
* Add granite template to test-chat-template.cpp
* Update src/llama.cpp
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* Update tests/test-chat-template.cpp
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* Added proper template and expected output
* Small change to \n
Small change to \n
* Add code space &
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* Fix spacing
* Apply suggestions from code review
* Update src/llama.cpp
---------
Co-authored-by: Xuan Son Nguyen <thichthat@gmail.com>
* [CANN] Adapt to dynamically loadable backends mechanism
* Fix the Bug: inference running result is garbled in debug running model for LM models who's type is Q4_0 class
* Handle the review comments of this pull request
* AVX512 version of ggml_gemm_q4_0_8x8_q8_0
* Remove zero vector parameter passing
* Rename functions and rearrange order of macros
* Edit commments
* style : minor adjustments
* Update x to start from 0
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* Add AVX2 based implementations for quantize_q8_0_4x8, ggml_gemv_q4_0_8x8_q8_0 and ggml_gemm_q4_0_8x8_q8_0 functions
* Update code to fix issues occuring due to non alignment of elements to be processed as multiple of 16 in MSVC
* Update comments and indentation
* Make updates to reduce number of load instructions
* chore: Fix compiler warnings, add help text, improve CLI options
* Add prototypes for function definitions
* Invert logic of --no-clean option to be more intuitive
* Provide a new help prompt with clear instructions
* chore : Add ignore rule for vulkan shader generator
Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>
* Update ggml/src/vulkan-shaders/vulkan-shaders-gen.cpp
Co-authored-by: 0cc4m <picard12@live.de>
* chore : Remove void and apply C++ style empty parameters
* chore : Remove void and apply C++ style empty parameters
---------
Signed-off-by: teleprint-me <77757836+teleprint-me@users.noreply.github.com>
Co-authored-by: 0cc4m <picard12@live.de>
Changes:
- Move each example into its own function. This makes the code much
easier to read and understand.
- Make the program easy to only run one test by commenting out function
calls in main().
- Make the output easy to parse by indenting the output for each example.
- Add shebang and +x bit to make it clear it's an executable.
- Make the host configurable via --host with a default 127.0.0.1:8080.
- Make the code look in the tools list to call the registered tool,
instead of hardcoding the returned values. This makes the code more
copy-pastable.
- Add error checking, so that the program exits 1 if the LLM didn't
returned expected values. It's super useful to check for correctness.
Testing:
- Tested with Mistral-7B-Instruct-v0.3 in F16 and Q5_K_M and
Meta-Llama-3-8B-Instruct in F16 and Q5_K_M.
- I did not observe a failure even once in Mistral-7B-Instruct-v0.3.
- Llama-3 failed about a third of the time in example_concurrent: it
only returned one call instead of 3. Even for F16.
Potential follow ups:
- Do not fix the prompt encoding yet. Surprisingly it mostly works even
if the prompt encoding is not model optimized.
- Add chained answer and response.
Test only change.
Xuan-Son Nguyen