MCPcopy Index your code

hub / github.com/ml-explore/mlx-lm / types & classes

Types & classes816 in github.com/ml-explore/mlx-lm

↓ 58 callersClassKVCache
mlx_lm/models/cache.py:325
↓ 41 callersClassArraysCache
mlx_lm/models/cache.py:594
↓ 38 callersClassRotatingKVCache
mlx_lm/models/cache.py:410
↓ 32 callersClassSwitchGLU
mlx_lm/models/switch_layers.py:160
↓ 28 callersClassMockCache
tests/test_server.py:75
↓ 18 callersClassBatchGenerator
A batch generator implements continuous batching. This class provides automatic management of prompt processing and generation batches,
mlx_lm/generate.py:1486
↓ 13 callersClassScaleConfig
mlx_lm/quant/awq.py:25
↓ 12 callersClassCacheList
mlx_lm/models/cache.py:814
↓ 10 callersClassMultiLinear
mlx_lm/models/mla.py:9
↓ 9 callersClassBatchRotatingKVCache
mlx_lm/models/cache.py:1133
↓ 9 callersClassLRUPromptCache
mlx_lm/models/cache.py:1623
↓ 8 callersClassBitLinear
BitLinear module with memory-efficient weight handling.
mlx_lm/models/bitlinear_layers.py:92
↓ 7 callersClassBatchKVCache
mlx_lm/models/cache.py:912
↓ 7 callersClassRMSNorm
mlx_lm/models/gemma3_text.py:104
↓ 7 callersClassSwitchLinear
mlx_lm/models/switch_layers.py:93
↓ 6 callersClassSequenceStateMachine
A state machine that uses one Aho-Corasick trie per state to efficiently track state across a generated sequence. The transitions are provide
mlx_lm/generate.py:943
↓ 5 callersClassRMSNorm
mlx_lm/models/gemma2.py:30
↓ 5 callersClassRMSNorm
mlx_lm/models/plamo2.py:40
↓ 5 callersClassZeroCenteredRMSNorm
mlx_lm/models/step3p5.py:66
↓ 4 callersClassLoRA
mlx_lm/models/rwkv7.py:162
↓ 4 callersClassPromptTrieResult
mlx_lm/models/cache.py:1524
↓ 4 callersClassSuScaledRoPE
mlx_lm/models/rope_utils.py:10
↓ 3 callersClassAWQConfig
mlx_lm/quant/awq.py:34
↓ 3 callersClassCacheDataset
mlx_lm/tuner/datasets.py:158
↓ 3 callersClassChunkedKVCache
mlx_lm/models/cache.py:731
↓ 3 callersClassDeepseekV3Model
mlx_lm/models/deepseek_v3.py:319
↓ 3 callersClassFusedLinear
mlx_lm/models/afm7.py:123
↓ 3 callersClassNemotronLayerNorm1P
mlx_lm/models/nemotron.py:54
↓ 3 callersClassQuantizedKVCache
mlx_lm/models/cache.py:232
↓ 3 callersClassRMSNorm
mlx_lm/models/recurrent_gemma.py:39
↓ 3 callersClassRMSNorm
mlx_lm/models/gemma.py:27
↓ 3 callersClassResponse
mlx_lm/server.py:226
↓ 3 callersClassResponseGenerator
mlx_lm/server.py:440
↓ 3 callersClassShortConv1d
mlx_lm/models/kimi_linear.py:235
↓ 2 callersClassAPIHandler
mlx_lm/server.py:1059
↓ 2 callersClassAttention
mlx_lm/models/gpt2.py:29
↓ 2 callersClassBailingMoeMLP
mlx_lm/models/bailing_moe.py:60
↓ 2 callersClassCompletionRequest
mlx_lm/server.py:198
↓ 2 callersClassDeepseekMLP
mlx_lm/models/deepseek.py:108
↓ 2 callersClassDeepseekV2MLP
mlx_lm/models/deepseek_v2.py:248
↓ 2 callersClassDeepseekV32MLP
mlx_lm/models/deepseek_v32.py:262
↓ 2 callersClassDeepseekV3MLP
mlx_lm/models/deepseek_v3.py:172
↓ 2 callersClassDoRALinear
mlx_lm/tuner/dora.py:9
↓ 2 callersClassDots1MLP
mlx_lm/models/dots1.py:162
↓ 2 callersClassDummyModelProvider
tests/test_server.py:24
↓ 2 callersClassErnie4_5_MLP
mlx_lm/models/ernie4_5_moe.py:94
↓ 2 callersClassFusedLoRALinear
mlx_lm/models/afm7.py:32
↓ 2 callersClassGenerationContext
mlx_lm/server.py:209
↓ 2 callersClassGenerationResponse
The output of :func:`stream_generate`. Args: text (str): The next segment of decoded text. This can be an empty string. toke
mlx_lm/generate.py:270
↓ 2 callersClassGlm4MoeLiteMLP
mlx_lm/models/glm4_moe_lite.py:177
↓ 2 callersClassKimiMLP
mlx_lm/models/kimi_linear.py:57
↓ 2 callersClassKlearMLP
mlx_lm/models/Klear.py:110
↓ 2 callersClassLayerNorm2D
mlx_lm/models/cohere.py:30
↓ 2 callersClassLayerNormPerHead
mlx_lm/models/stablelm.py:30
↓ 2 callersClassLinearSubblockReplacement
A simple linear layer used to replace Attention or MLP blocks.
mlx_lm/models/nemotron-nas.py:238
↓ 2 callersClassLongcatFlashDecoderLayer
mlx_lm/models/longcat_flash.py:280
↓ 2 callersClassMLP
mlx_lm/models/bailing_moe_linear.py:114
↓ 2 callersClassMLP
mlx_lm/models/gpt2.py:71
↓ 2 callersClassMLP
mlx_lm/models/glm4_moe.py:111
↓ 2 callersClassMLP
mlx_lm/models/mimo_v2_flash.py:127
↓ 2 callersClassMLP
mlx_lm/models/exaone_moe.py:113
↓ 2 callersClassMLP
mlx_lm/models/afm7.py:266
↓ 2 callersClassMLP
mlx_lm/models/llama4.py:137
↓ 2 callersClassMLP
mlx_lm/models/hunyuan.py:144
↓ 2 callersClassMLP
mlx_lm/models/afmoe.py:137
↓ 2 callersClassMLXLM
mlx_lm/evaluate.py:72
↓ 2 callersClassNemotronHMLP
mlx_lm/models/nemotron_h.py:297
↓ 2 callersClassQwen3NextMLP
mlx_lm/models/qwen3_next.py:161
↓ 2 callersClassSparseMoeBlock
mlx_lm/models/dbrx.py:125
↓ 2 callersClassStep3p5MLP
mlx_lm/models/step3p5.py:76
↓ 2 callersClassSwiGLU
mlx_lm/models/gpt_oss.py:62
↓ 2 callersClassSwitchMLP
mlx_lm/models/switch_layers.py:202
↓ 2 callersClassTokenShift
mlx_lm/models/rwkv7.py:198
↓ 2 callersClassTokenizerWrapper
A wrapper that combines an HF tokenizer and a detokenizer. Accessing any attribute other than the ``detokenizer`` is forwarded to the hugging
mlx_lm/tokenizer_utils.py:287
↓ 2 callersClassTrainingArgs
mlx_lm/tuner/trainer.py:42
↓ 1 callersClassAFMModel
mlx_lm/models/afm7.py:330
↓ 1 callersClassAfmoeMoE
mlx_lm/models/afmoe.py:167
↓ 1 callersClassAfmoeModel
mlx_lm/models/afmoe.py:278
↓ 1 callersClassApertusAttention
mlx_lm/models/apertus.py:51
↓ 1 callersClassApertusDecoderLayer
mlx_lm/models/apertus.py:117
↓ 1 callersClassApertusMLP
mlx_lm/models/apertus.py:36
↓ 1 callersClassApertusModel
mlx_lm/models/apertus.py:137
↓ 1 callersClassAttention
mlx_lm/models/qwen2.py:32
↓ 1 callersClassAttention
mlx_lm/models/ernie4_5.py:31
↓ 1 callersClassAttention
mlx_lm/models/cohere2.py:33
↓ 1 callersClassAttention
mlx_lm/models/gemma4_text.py:176
↓ 1 callersClassAttention
mlx_lm/models/bailing_moe_linear.py:137
↓ 1 callersClassAttention
mlx_lm/models/lfm2_moe.py:54
↓ 1 callersClassAttention
mlx_lm/models/qwen3_moe.py:37
↓ 1 callersClassAttention
mlx_lm/models/openelm.py:57
↓ 1 callersClassAttention
mlx_lm/models/plamo.py:28
↓ 1 callersClassAttention
mlx_lm/models/granite.py:36
↓ 1 callersClassAttention
mlx_lm/models/glm4_moe.py:49
↓ 1 callersClassAttention
mlx_lm/models/mimo_v2_flash.py:54
↓ 1 callersClassAttention
mlx_lm/models/gemma.py:37
↓ 1 callersClassAttention
mlx_lm/models/exaone_moe.py:164
↓ 1 callersClassAttention
mlx_lm/models/minicpm.py:45
↓ 1 callersClassAttention
mlx_lm/models/afm7.py:165
↓ 1 callersClassAttention
mlx_lm/models/llama.py:45
↓ 1 callersClassAttention
mlx_lm/models/minicpm3.py:39
next →1–100 of 816, ranked by callers