Code
Hub
Workspaces
Following
Trending
Connect
MCP
copy
Create free account
hub
/
github.com/UbiquitousLearning/mllm
/ functions
Functions
5,971 in github.com/UbiquitousLearning/mllm
⨍
Functions
5,971
◇
Types & classes
1,855
↳
Endpoints
4
↓ 1 callers
Function
processIntrusivePtr
Benchmark test: function passing and returning
benchmarks/ext_stl/intrusive_ptr.cpp:305
↓ 1 callers
Function
processSharedPtr
benchmarks/ext_stl/intrusive_ptr.cpp:310
↓ 1 callers
Method
ptr
mllm/core/Tensor.hpp:646
↓ 1 callers
Method
purge
mllm/engine/prefix_cache/ZenFS.cpp:141
↓ 1 callers
Function
qhmath_hvx_dequantize_add_af
Only support 128x dimension
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantizeAdd.cpp:276
↓ 1 callers
Function
qhmath_hvx_dequantize_add_ahf
execute functions for ops */
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantizeAdd.cpp:108
↓ 1 callers
Function
qhmath_hvx_dequantize_add_ui16_af
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantizeAdd.cpp:381
↓ 1 callers
Function
qhmath_hvx_dequantize_add_ui16_ahf
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantizeAdd.cpp:200
↓ 1 callers
Function
qhmath_hvx_dequantize_af
Only support 128x dimension
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantize.cpp:221
↓ 1 callers
Function
qhmath_hvx_dequantize_ahf
execute functions for ops */
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantize.cpp:107
↓ 1 callers
Function
qhmath_hvx_dequantize_ui16_af
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantize.cpp:301
↓ 1 callers
Function
qhmath_hvx_dequantize_ui16_ahf
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMADequantize.cpp:168
↓ 1 callers
Function
qhmath_hvx_quantize_af
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:857
↓ 1 callers
Function
qhmath_hvx_quantize_af_out_int8
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:1098
↓ 1 callers
Function
qhmath_hvx_quantize_ahf
execute functions for ops */
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:113
↓ 1 callers
Function
qhmath_hvx_quantize_ahf_int8
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:604
↓ 1 callers
Function
qhmath_hvx_quantize_ui16_af
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:979
↓ 1 callers
Function
qhmath_hvx_quantize_ui16_ahf
mllm/backends/qnn/custom-op-package/LLaMAPackage/src/ops/LLaMAQuantize.cpp:359
↓ 1 callers
Function
qnnDataTypeToSize
mllm/backends/qnn/QNNUtils.cpp:399
↓ 1 callers
Function
qnn_
()
pymllm/ffi/__init__.py:114
↓ 1 callers
Function
quantize_q8_0_4x4
mllm/backends/cpu/kernels/arm/linear/ggml/gemm_aarch64.cpp:97
↓ 1 callers
Function
quantize_q8_0_4x8
mllm/backends/cpu/kernels/arm/linear/ggml/gemm_aarch64.cpp:191
↓ 1 callers
Function
quantize_row_i8_reference
========================== smoothquant i8 =================================
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q8.cpp:264
↓ 1 callers
Function
quantize_row_iq2_xxs_impl
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q2.cpp:323
↓ 1 callers
Function
quantize_row_q2_K
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q2.cpp:172
↓ 1 callers
Function
quantize_row_q2_K_ref
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q2.cpp:106
↓ 1 callers
Function
quantize_row_q3_K
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q3.cpp:168
↓ 1 callers
Function
quantize_row_q3_K_ref
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q3.cpp:93
↓ 1 callers
Function
quantize_row_q4_0_reference
reference implementation for deterministic creation of model files
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q4.cpp:33
↓ 1 callers
Function
quantize_row_q4_K
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q4.cpp:325
↓ 1 callers
Function
quantize_row_q4_K_reference
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q4.cpp:184
↓ 1 callers
Function
quantize_row_q6_K
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q6.cpp:153
↓ 1 callers
Function
quantize_row_q6_K_reference
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q6.cpp:34
↓ 1 callers
Function
quantize_row_q8_0
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q8.cpp:57
↓ 1 callers
Function
quantize_row_q8_0_reference
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q8.cpp:32
↓ 1 callers
Function
quantize_row_q8_K
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q8.cpp:259
↓ 1 callers
Function
quantize_row_q8_K_reference
mllm/backends/cpu/kernels/common/ggml/quantize/quantize_q8.cpp:215
↓ 1 callers
Function
quick_gelu_fp16
mllm/backends/cpu/kernels/arm/gelu.cpp:207
↓ 1 callers
Function
qwen2VLRegex
mllm/models/qwen2vl/tokenization_qwen2vl.hpp:137
↓ 1 callers
Function
qwen2VLTokenizerMatchPattern
we need to handle this: (?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+
mllm/models/qwen2vl/tokenization_qwen2vl.hpp:20
↓ 1 callers
Function
qwen2vlLlmPartBench
playground/cache_reduced_computation/main.cpp:17
↓ 1 callers
Function
qwen3Bench
playground/cache_reduced_computation/main.cpp:37
↓ 1 callers
Function
qwen3Regex
mllm/models/qwen3/tokenization_qwen3.hpp:136
↓ 1 callers
Function
qwen3TokenizerMatchPattern
we need to handle this: (?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}| ?[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|\s+(?!\S)|\s+
mllm/models/qwen3/tokenization_qwen3.hpp:19
↓ 1 callers
Function
qwenRegex
mllm/models/qwen_npu/tokenization_qwen.hpp:130
↓ 1 callers
Function
qwenTokenizerMatchPattern
we need to handle this regex pattern similar to qwen3: (?i:'s|'t|'re|'ve|'m|'ll|'d)|[^\r\n\p{L}\p{N}]?\p{L}+|\p{N}|[^\s\p{L}\p{N}]+[\r\n]*|\s*[\r\n]+|
mllm/models/qwen_npu/tokenization_qwen.hpp:19
↓ 1 callers
Function
radixAttnSWAwSink
mllm-ext-opset/cpu/radix_swa_sink/tests/main.cpp:8
↓ 1 callers
Method
random
mllm/core/Tensor.cpp:140
↓ 1 callers
Function
readFeats
mllm/preprocessor/audio/Audio.cpp:130
↓ 1 callers
Method
receive
mllm/engine/Dispatcher.cpp:16
↓ 1 callers
Method
recompute_scale_zp
(self)
pymllm/backends/qualcomm/transformers/qwen3/runner.py:350
↓ 1 callers
Method
recompute_scale_zp
(self)
pymllm/backends/qualcomm/transformers/qwen2/runner.py:336
↓ 1 callers
Method
recompute_scale_zp
(self)
pymllm/backends/qualcomm/transformers/llama/runner.py:329
↓ 1 callers
Function
recursion_dump_cls
(father_name: str, cls: Cls, rets: List[str])
mllm/compile/ir/rtti_kind_gen.py:133
↓ 1 callers
Function
recursiveAttachGraphNameAndContextName
mllm/backends/qnn/aot/passes/SplitLLMGraphPass.cpp:20
↓ 1 callers
Function
recursiveCheckConcatInputs
mllm/backends/qnn/aot/passes/PTQPass.cpp:371
↓ 1 callers
Function
recursiveCheckUnsolved
mllm/backends/qnn/aot/passes/PTQPass.cpp:330
↓ 1 callers
Function
recursiveIndexTensorOp
mllm/compile/passes/EagerMemorySolverPass.cpp:18
↓ 1 callers
Function
recursiveInsertFreeTensorOp
mllm/compile/passes/EagerMemorySolverPass.cpp:41
↓ 1 callers
Function
recursiveRegisterLinalgSymbolOps
mllm/compile/passes/ExtractSymbolsPass.cpp:14
↓ 1 callers
Function
recursiveRemoveDbgInfo
mllm/compile/passes/EliminateDbgInfoPass.cpp:13
↓ 1 callers
Function
recursiveRemoveOpsIntoNewGraph
mllm/backends/qnn/aot/passes/SplitLLMGraphPass.cpp:37
↓ 1 callers
Function
recursiveSolveNormal
mllm/backends/qnn/aot/passes/PTQPass.cpp:308
↓ 1 callers
Function
recursiveSolveWeights
mllm/backends/qnn/aot/passes/PTQPass.cpp:137
↓ 1 callers
Function
recursiveVisitGraph
mllm/backends/qnn/aot/passes/LLMQuantRecipePass.cpp:22
↓ 1 callers
Function
recursiveVisitGraph
mllm/backends/qnn/aot/passes/MarkQnnGraphPass.cpp:24
↓ 1 callers
Method
refSessionThreads
mllm/engine/Context.cpp:140
↓ 1 callers
Function
reflect_index
Helper: reflect index to [0, size-1] without repeating edge (PyTorch-like reflect)
mllm/backends/cpu/ops/PadOp.cpp:28
↓ 1 callers
Method
regOpFactory
mllm/backends/base/Backend.cpp:15
↓ 1 callers
Method
registerCustomizedOp
mllm/backends/base/PluginSystem.cpp:82
↓ 1 callers
Method
registerQnnTensorToSharedBuffer
mllm/backends/qnn/QNNAllocator.cpp:79
↓ 1 callers
Function
register_jit_kernel
Register one JIT kernel metadata entry into the global table.
mllm-kernel/mllm_kernel/jit_utils/compile.py:212
↓ 1 callers
Method
register_pass
(self, pass_: QuantizeBasePass)
pymllm/quantize/solver.py:19
↓ 1 callers
Function
relu_fp16
mllm/backends/cpu/kernels/arm/relu.cpp:86
↓ 1 callers
Function
relu_fp32
mllm/backends/cpu/kernels/arm/relu.cpp:11
↓ 1 callers
Method
report
mllm/engine/BuddyMemPool.cpp:99
↓ 1 callers
Method
reset
mllm/utils/IntrusivePtr.hpp:93
↓ 1 callers
Method
reshape
mllm/backends/opencl/ops/LinearOp.cpp:147
↓ 1 callers
Method
reshape
mllm/backends/cpu/ops/LinearOp.cpp:319
↓ 1 callers
Method
reshape
mllm/backends/ascend/ops/AscendSliceOp.cpp:26
↓ 1 callers
Method
reshape
mllm/backends/qnn/CustomLayers.cpp:58
↓ 1 callers
Method
retain
mllm/backends/opencl/vendors/OpenCL-Headers/CL/opencl.hpp:2268
↓ 1 callers
Method
retrieveContext
mllm/backends/qnn/QNNBackend.cpp:408
↓ 1 callers
Function
rmsNorm
mllm/nn/Functional.cpp:141
↓ 1 callers
Function
rmsnorm_fp16
Should support [B, S, H * D] and [B, S, H, D]
mllm/backends/cpu/kernels/arm/rmsnorm.cpp:101
↓ 1 callers
Function
rmsnorm_fp16_inplace
mllm/backends/cpu/kernels/arm/rmsnorm.cpp:139
↓ 1 callers
Function
rmsnorm_fp32_inplace
mllm/backends/cpu/kernels/arm/rmsnorm.cpp:56
↓ 1 callers
Method
root_cause
mllm-kernel/include/mllm_kernel/utils.hpp:71
↓ 1 callers
Function
round_ste
Round with straight-through estimator
mllm/models/minicpm_o2_6/modeling_vector_quantize.hpp:21
↓ 1 callers
Method
rparentheses
mllm/compile/ir/IRPrinter.hpp:33
↓ 1 callers
Method
rsbracket
mllm/compile/ir/IRPrinter.hpp:37
↓ 1 callers
Function
save
mllm/mllm.cpp:146
↓ 1 callers
Method
save
Saves the current JSON object to a file.
mllm/engine/ConfigFile.cpp:47
↓ 1 callers
Method
select_high_score_visual_token_decode
algorithms/lazy_vlm/models/qwen2vl/lazy_vlm_cfg.hpp:127
↓ 1 callers
Method
select_high_score_visual_token_prefill
algorithms/lazy_vlm/models/qwen2vl/lazy_vlm_cfg.hpp:87
↓ 1 callers
Function
send_request
(request: str)
pymllm/service/rr_process.py:18
↓ 1 callers
Method
setBlockwiseQuantization
mllm/backends/qnn/QNNUtils.cpp:517
↓ 1 callers
Method
setConfig
mllm/backends/qnn/aot/passes/AOTCompileContext.cpp:12
↓ 1 callers
Method
setCpuOpThreads
mllm/engine/Context.cpp:161
↓ 1 callers
Method
setDeviceType
mllm/core/BaseOp.cpp:22
← previous
next →
1,601–1,700 of 5,971, ranked by callers