MCPcopy Create free account

hub / github.com/UbiquitousLearning/mllm / functions

Functions5,971 in github.com/UbiquitousLearning/mllm

⨍ Functions 5,971◇ Types & classes 1,855↳ Endpoints 4
↓ 2 callersMethodallocKey
mllm/nn/lmcache/PrefixCache.cpp:24
↓ 2 callersMethodallocValue
mllm/nn/lmcache/PrefixCache.cpp:26
↓ 2 callersFunctionaopsFromJson
mllm/compile/jit/interpreter/AopsFromJson.cpp:51
↓ 2 callersMethodapplyChatTemplate
mllm/models/qwen3/modeling_qwen3_service.hpp:559
↓ 2 callersFunctionasciiTitle
Generate the specified ASCII art title "MLLM TEAM"
mllm-cli/tui/welcome.go:23
↓ 2 callersFunctionbatch_matmul_fp32
mllm/backends/cpu/kernels/common/blas.cpp:33
↓ 2 callersFunctionbitReverse
mllm/backends/cpu/ops/STFTOp.cpp:68
↓ 2 callersFunctioncalculateBroadcastShape
Helper function to calculate broadcast shape
mllm/backends/cpu/ops/MaskedScatterOp.cpp:16
↓ 2 callersFunctioncall_elewise_add_fp32
mllm/backends/cpu/kernels/common/kernel_dispatch.cpp:37
↓ 2 callersFunctioncall_elewise_div_fp32
mllm/backends/cpu/kernels/common/kernel_dispatch.cpp:49
↓ 2 callersFunctioncall_elewise_mul_fp32
mllm/backends/cpu/kernels/common/kernel_dispatch.cpp:45
↓ 2 callersFunctioncall_elewise_sub_fp32
mllm/backends/cpu/kernels/common/kernel_dispatch.cpp:41
↓ 2 callersMethodcategory_flag
mllm/preprocessor/tokenizers/llama_cpp_unicode/unicode.h:69
↓ 2 callersFunctioncheck_llamafile_sgemm
mllm/backends/cpu/kernels/common/llamafile/llamafile_sgemm.cpp:958
↓ 2 callersFunctioncl_half_handle_overflow
* Utility to deal with values that overflow when converting to half precision. */
mllm/backends/opencl/vendors/OpenCL-Headers/CL/cl_half.h:61
↓ 2 callersFunctioncl_half_handle_underflow
* Utility to deal with values that underflow when converting to half precision. */
mllm/backends/opencl/vendors/OpenCL-Headers/CL/cl_half.h:87
↓ 2 callersMethodclip
mllm/core/Tensor.cpp:402
↓ 2 callersFunctionclip_sampler
mllm/preprocessor/audio/Audio.cpp:368
↓ 2 callersMethodcompile
(self)
pymllm/backends/qualcomm/transformers/qwen3/runner.py:230
↓ 2 callersMethodcompile
mllm/backends/qnn/aot/QnnWrappersAPI.cpp:344
↓ 2 callersFunctioncompute_strides_internal
mllm/backends/cpu/kernels/common/permute-inl.hpp:10
↓ 2 callersMethodconvert_tokens_to_ids
mllm/preprocessor/tokenizers/WordPiece.cpp:176
↓ 2 callersFunctioncopy
mllm/engine/io/CpuMemoryDiskDispatcher.cpp:12
↓ 2 callersMethodcreateContext
mllm/backends/qnn/QNNBackend.cpp:399
↓ 2 callersFunctioncreateIRTraceDispatcher
mllm/compile/ir/IRTraceDispatcher.cpp:51
↓ 2 callersFunctioncreateLLMCanonicalizationPipeline
mllm/compile/passes/LLMCanonicalizationPipeline.cpp:11
↓ 2 callersFunctioncreateQNNGraphBuildPass
mllm/backends/qnn/passes/QNNGraphBuildPass.hpp:41
↓ 2 callersFunctioncreateQNNGraphIOTensorPass
mllm/backends/qnn/passes/QNNGraphIOTensorPass.cpp:87
↓ 2 callersFunctioncreateQNNOpNamingPass
mllm/backends/qnn/passes/QNNOpNamingPass.cpp:101
↓ 2 callersFunctioncreateQuantizeParams
mllm/backends/qnn/QNNUtils.cpp:605
↓ 2 callersFunctioncreate_session
(model_name: str)
pymllm/service/models_hub.py:21
↓ 2 callersMethodcrop
mllm/preprocessor/visual/Image.cpp:109
↓ 2 callersMethodctrlByMemManager
mllm/backends/cpu/CPUAllocator.hpp:13
↓ 2 callersFunctioncubic_interp1d
Bicubic interpolation helper function
mllm/backends/cpu/ops/InterpolateOp.cpp:179
↓ 2 callersMethodcurTime
mllm/engine/Context.cpp:134
↓ 2 callersMethoddecode
mllm/models/minicpm_o2_6/tokenization_chattts.cpp:39
↓ 2 callersMethoddecode
mllm/models/deepseek_ocr/tokenization_deepseek_ocr.hpp:59
↓ 2 callersMethoddot
mllm/nn/lmcache/PrefixCache.cpp:16
↓ 2 callersFunctiondownload_mllm_model
Download a model from the model hub. Args: model_name: Name of the model to download model_dir: Directory to save the model
pymllm/service/models_hub.py:30
↓ 2 callersFunctiondynamicPreprocess
* Dynamic preprocess that crops and resizes an image into tiles matching * a target aspect ratio grid, similar to DeepSeek-OCR implementation. * *
mllm/models/deepseek_ocr/conversation_preprocess.hpp:400
↓ 2 callersMethodenqueueCopyBuffer
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:8103
↓ 2 callersMethodenqueueMapBuffer
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:8695
↓ 2 callersFunctionenqueueMapSVM
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:10111
↓ 2 callersMethodenqueueReadBuffer
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:8053
↓ 2 callersMethodenqueueUnmapMemObject
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:9014
↓ 2 callersFunctionew_add_fp32
mllm/backends/cpu/kernels/arm/elementwise.cpp:14
↓ 2 callersFunctionew_div_fp32
mllm/backends/cpu/kernels/arm/elementwise.cpp:32
↓ 2 callersFunctionew_mul_fp32
mllm/backends/cpu/kernels/arm/elementwise.cpp:26
↓ 2 callersFunctionew_sub_fp32
mllm/backends/cpu/kernels/arm/elementwise.cpp:20
↓ 2 callersMethodfile_name
mllm-kernel/include/mllm_kernel/source_location.hpp:32
↓ 2 callersFunctionfill_arange
mllm/backends/cpu/kernels/arm/fill.cpp:54
↓ 2 callersFunctionfill_arange_fp16
mllm/backends/cpu/kernels/arm/fill.cpp:131
↓ 2 callersFunctionfill_ones
mllm/backends/cpu/kernels/arm/fill.cpp:26
↓ 2 callersFunctionfill_ones_fp16
mllm/backends/cpu/kernels/arm/fill.cpp:103
↓ 2 callersFunctionfill_random
mllm/backends/cpu/kernels/arm/fill.cpp:72
↓ 2 callersFunctionfill_random_fp16
mllm/backends/cpu/kernels/arm/fill.cpp:156
↓ 2 callersFunctionfill_specific_value
mllm/backends/cpu/kernels/arm/fill.cpp:40
↓ 2 callersFunctionfill_specific_value_fp16
mllm/backends/cpu/kernels/arm/fill.cpp:117
↓ 2 callersFunctionfill_zeros
mllm/backends/cpu/kernels/arm/fill.cpp:12
↓ 2 callersFunctionfill_zeros_fp16
mllm/backends/cpu/kernels/arm/fill.cpp:89
↓ 2 callersMethodfinalizeGraph
mllm/backends/qnn/QNNModel.cpp:307
↓ 2 callersMethodfinish
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:9578
↓ 2 callersFunctionfloat_to_bits
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantizeAdd.cpp:99
↓ 2 callersFunctionfloat_to_bits
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantize.cpp:98
↓ 2 callersFunctionfloat_to_bits
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/SiLU.cpp:97
↓ 2 callersFunctionfloat_to_fp16s
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/RMSNorm.cpp:199
↓ 2 callersMethodfloor
mllm/compile/symbolic_expr/Eval.cpp:25
↓ 2 callersFunctionfork
mllm/mllm.hpp:79
↓ 2 callersMethodforward
mllm/backends/opencl/ops/X2XOp.cpp:20
↓ 2 callersMethodforward
mllm/backends/cpu/ops/X2XOp.cpp:11
↓ 2 callersMethodforward
mllm/backends/ascend/ops/AscendX2XOp.cpp:17
↓ 2 callersMethodforward
mllm/backends/qnn/op/QNNX2XOp.cpp:13
↓ 2 callersMethodforward
mllm/nn/Module.hpp:216
↓ 2 callersMethodforward
mllm/models/minicpm_o2_6/audio_preprocessor_minicpmo.hpp:167
↓ 2 callersMethodfreeCachedTensors
mllm/backends/qnn/QNNModel.cpp:328
↓ 2 callersMethodgeneralAlloc
mllm/backends/cpu/CPUAllocator.cpp:57
↓ 2 callersFunctionget
mllm/utils/Enumerate.hpp:209
↓ 2 callersFunctionget2DSinCosPosEmbed
mllm/models/minicpm_o2_6/modeling_resampler.hpp:21
↓ 2 callersMethodgetAutoIndexedValueName
mllm/compile/ir/Node.cpp:230
↓ 2 callersMethodgetCurrentSeqCnt
mllm/backends/cpu/ops/KVCacheOp.cpp:69
↓ 2 callersFunctiongetDevicePlatformVersion
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:2174
↓ 2 callersMethodgetKVCache
algorithms/lazy_vlm/HKVCacheFast.cpp:79
↓ 2 callersFunctiongetMultiDimIndices
Helper function to convert linear index to multi-dimensional indices
mllm/backends/cpu/ops/MaskedScatterOp.cpp:61
↓ 2 callersMethodgetNativeParam
mllm/backends/qnn/QNNUtils.cpp:594
↓ 2 callersMethodgetOutputFromIndices
* Get output from indices by summing across quantizers * @param indices Quantized indices * @return Reconstructed output tensor */
mllm/models/minicpm_o2_6/modeling_vector_quantize.hpp:657
↓ 2 callersMethodgetParams
mllm/core/aops/ParamOp.cpp:49
↓ 2 callersMethodgetPrecedence
mllm/compile/symbolic_expr/Parser.cpp:117
↓ 2 callersMethodgetQnnGraph
mllm/backends/qnn/QNNModel.hpp:69
↓ 2 callersMethodgetRandomState
mllm/engine/Context.cpp:127
↓ 2 callersFunctiongetRuntimeTypeTraits
mllm/backends/cpu/kernels/common/ggml/vec_dot_type.hpp:155
↓ 2 callersMethodgetThreadPool
mllm/backends/cpu/CPUBackend.cpp:93
↓ 2 callersFunctionget_download_model_path
(model_name_or_model_path: str)
pymllm/service/models_hub.py:65
↓ 2 callersFunctionget_scale_min_k4
Helper function for Q4_K quantization
mllm/backends/cpu/kernels/common/ggml/vec_dot_type.cpp:36
↓ 2 callersFunctionget_scale_shuffle_k4
mllm/backends/cpu/kernels/common/ggml/quantize/quantize.hpp:237
↓ 2 callersMethodget_shell_context
Get a persistent shell context to a device. This allows running commands that maintain state (e.g., environment variables). I
pymllm/utils/adb.py:206
↓ 2 callersFunctionget_wav_data
mllm/preprocessor/audio/Audio.cpp:355
↓ 2 callersMethodgraphFinalize
mllm/backends/qnn/QNNBackend.cpp:617
↓ 2 callersFunctionhasAVX
mllm/utils/CPUArchHelper.hpp:261
↓ 2 callersFunctionhasAVX2
mllm/utils/CPUArchHelper.hpp:267
↓ 2 callersFunctionhasAVX512BW
mllm/utils/CPUArchHelper.hpp:279
← previousnext →601–700 of 5,971, ranked by callers