bigdl-core-cpp 2.6.0b20250320__py3-none-win_amd64.whl → 2.6.0b20250321__py3-none-win_amd64.whl

bigdl/cpp/gguf-py/gguf/gguf_writer.py

@@ -26,6 +26,7 @@ from .constants import (
     RopeScalingType,
     PoolingType,
     TokenType,
+    ExpertGatingFuncType,
 )
 
 from .quants import quant_shape_from_byte_shape
@@ -631,6 +632,21 @@ class GGUFWriter:
     def add_embedding_length(self, length: int) -> None:
         self.add_uint32(Keys.LLM.EMBEDDING_LENGTH.format(arch=self.arch), length)
 
+    def add_features_length(self, length: int) -> None:
+        self.add_uint32(Keys.LLM.FEATURES_LENGTH.format(arch=self.arch), length)
+
+    def add_posnet_embedding_length(self, length: int) -> None:
+        self.add_uint32(Keys.PosNet.EMBEDDING_LENGTH.format(arch=self.arch), length)
+
+    def add_posnet_block_count(self, length: int) -> None:
+        self.add_uint32(Keys.PosNet.BLOCK_COUNT.format(arch=self.arch), length)
+
+    def add_convnext_embedding_length(self, length: int) -> None:
+        self.add_uint32(Keys.ConvNext.EMBEDDING_LENGTH.format(arch=self.arch), length)
+
+    def add_convnext_block_count(self, length: int) -> None:
+        self.add_uint32(Keys.ConvNext.BLOCK_COUNT.format(arch=self.arch), length)
+
     def add_block_count(self, length: int) -> None:
         self.add_uint32(Keys.LLM.BLOCK_COUNT.format(arch=self.arch), length)
 
@@ -700,6 +716,12 @@ class GGUFWriter:
     def add_expert_weights_scale(self, value: float) -> None:
         self.add_float32(Keys.LLM.EXPERT_WEIGHTS_SCALE.format(arch=self.arch), value)
 
+    def add_expert_weights_norm(self, value: bool) -> None:
+        self.add_bool(Keys.LLM.EXPERT_WEIGHTS_NORM.format(arch=self.arch), value)
+
+    def add_expert_gating_func(self, value: ExpertGatingFuncType) -> None:
+        self.add_uint32(Keys.LLM.EXPERT_GATING_FUNC.format(arch=self.arch), value.value)
+
     def add_swin_norm(self, value: bool) -> None:
         self.add_bool(Keys.LLM.SWIN_NORM.format(arch=self.arch), value)
 
@@ -721,12 +743,21 @@ class GGUFWriter:
     def add_wkv_head_size(self, size: int) -> None:
         self.add_uint32(Keys.WKV.HEAD_SIZE.format(arch=self.arch), size)
 
+    def add_token_shift_count(self, count: int) -> None:
+        self.add_uint32(Keys.LLM.TOKEN_SHIFT_COUNT.format(arch=self.arch), count)
+
     def add_layer_norm_eps(self, value: float) -> None:
         self.add_float32(Keys.Attention.LAYERNORM_EPS.format(arch=self.arch), value)
 
     def add_layer_norm_rms_eps(self, value: float) -> None:
         self.add_float32(Keys.Attention.LAYERNORM_RMS_EPS.format(arch=self.arch), value)
 
+    def add_group_norm_eps(self, value: float) -> None:
+        self.add_float32(Keys.Attention.GROUPNORM_EPS.format(arch=self.arch), value)
+
+    def add_group_norm_groups(self, value: int) -> None:
+        self.add_uint32(Keys.Attention.GROUPNORM_GROUPS.format(arch=self.arch), value)
+
     def add_causal_attention(self, value: bool) -> None:
         self.add_bool(Keys.Attention.CAUSAL.format(arch=self.arch), value)
 
@@ -826,9 +857,6 @@ class GGUFWriter:
     def add_pad_token_id(self, id: int) -> None:
         self.add_uint32(Keys.Tokenizer.PAD_ID, id)
 
-    def add_cls_token_id(self, id: int) -> None:
-        self.add_uint32(Keys.Tokenizer.CLS_ID, id)
-
     def add_mask_token_id(self, id: int) -> None:
         self.add_uint32(Keys.Tokenizer.MASK_ID, id)
 

bigdl/cpp/gguf-py/gguf/tensor_mapping.py

@@ -13,7 +13,7 @@ class TensorNameMap:
             "transformer.wte",                           # gpt2 gpt-j mpt refact qwen dbrx jais exaone
             "transformer.word_embeddings",               # falcon
             "word_embeddings",                           # bloom
-            "model.embed_tokens",                        # llama-hf nemotron olmoe olmo2
+            "model.embed_tokens",                        # llama-hf nemotron olmoe olmo2 rwkv6qwen2
             "tok_embeddings",                            # llama-pth
             "embeddings.word_embeddings",                # bert nomic-bert
             "language_model.embedding.word_embeddings",  # persimmon
@@ -42,6 +42,7 @@ class TensorNameMap:
             "emb_ln",                     # nomic-bert
             "transformer.norm",           # openelm
             "rwkv.blocks.0.pre_ln",       # rwkv
+            "backbone.norm",              # wavtokenizer
         ),
 
         # Position embeddings
@@ -54,19 +55,20 @@ class TensorNameMap:
         # Output
         MODEL_TENSOR.OUTPUT: (
             "embed_out",                 # gptneox
-            "lm_head",                   # gpt2 mpt falcon llama-hf baichuan qwen mamba dbrx jais nemotron exaone olmoe olmo2
+            "lm_head",                   # gpt2 mpt falcon llama-hf baichuan qwen mamba dbrx jais nemotron exaone olmoe olmo2 phimoe
             "output",                    # llama-pth bloom internlm2
             "word_embeddings_for_head",  # persimmon
             "lm_head.linear",            # phi2
             "output_layer",              # chatglm
             "head",                      # rwkv
+            "head.out",                  # wavtokenizer
         ),
 
         # Output norm
         MODEL_TENSOR.OUTPUT_NORM: (
             "gpt_neox.final_layer_norm",               # gptneox
             "transformer.ln_f",                        # gpt2 gpt-j falcon jais exaone
-            "model.norm",                              # llama-hf baichuan internlm2 olmoe olmo2
+            "model.norm",                              # llama-hf baichuan internlm2 olmoe olmo2 phimoe
             "norm",                                    # llama-pth
             "transformer.norm_f",                      # mpt dbrx
             "ln_f",                                    # refact bloom qwen gpt2
@@ -80,6 +82,7 @@ class TensorNameMap:
             "transformer.norm",                        # openelm
             "model.norm",                              # nemotron
             "rwkv.ln_out",                             # rwkv
+            "backbone.final_layer_norm",               # wavtokenizer
         ),
 
         # Rope frequencies
@@ -90,6 +93,13 @@ class TensorNameMap:
 
         MODEL_TENSOR.ROPE_FACTORS_LONG: (),
         MODEL_TENSOR.ROPE_FACTORS_SHORT: (),
+
+        MODEL_TENSOR.CONV1D: (
+            "backbone.embed", # roberta
+        ),
+
+        MODEL_TENSOR.ROPE_FACTORS_LONG: (),
+        MODEL_TENSOR.ROPE_FACTORS_SHORT: (),
     }
 
     block_mappings_cfg: dict[MODEL_TENSOR, tuple[str, ...]] = {
@@ -101,7 +111,7 @@ class TensorNameMap:
             "transformer.h.{bid}.input_layernorm",                  # falcon7b
             "h.{bid}.input_layernorm",                              # bloom
             "transformer.h.{bid}.ln_mlp",                           # falcon40b
-            "model.layers.{bid}.input_layernorm",                   # llama-hf nemotron olmoe
+            "model.layers.{bid}.input_layernorm",                   # llama-hf nemotron olmoe phimoe
             "layers.{bid}.attention_norm",                          # llama-pth
             "language_model.encoder.layers.{bid}.input_layernorm",  # persimmon
             "model.layers.{bid}.ln1",                               # yi
@@ -145,7 +155,7 @@ class TensorNameMap:
 
         # Attention query
         MODEL_TENSOR.ATTN_Q: (
-            "model.layers.{bid}.self_attn.q_proj",                       # llama-hf nemotron olmoe olmo2
+            "model.layers.{bid}.self_attn.q_proj",                       # llama-hf nemotron olmoe olmo2 phimoe
             "model.layers.{bid}.self_attn.q_proj_no_perm",               # llama-custom
             "layers.{bid}.attention.wq",                                 # llama-pth
             "encoder.layer.{bid}.attention.self.query",                  # bert
@@ -158,7 +168,7 @@ class TensorNameMap:
 
         # Attention key
         MODEL_TENSOR.ATTN_K: (
-            "model.layers.{bid}.self_attn.k_proj",                     # llama-hf nemotron olmoe olmo2
+            "model.layers.{bid}.self_attn.k_proj",                     # llama-hf nemotron olmoe olmo2 phimoe
             "model.layers.{bid}.self_attn.k_proj_no_perm",             # llama-custom
             "layers.{bid}.attention.wk",                               # llama-pth
             "encoder.layer.{bid}.attention.self.key",                  # bert
@@ -172,7 +182,7 @@ class TensorNameMap:
 
         # Attention value
         MODEL_TENSOR.ATTN_V: (
-            "model.layers.{bid}.self_attn.v_proj",                       # llama-hf nemotron olmoe olmo2
+            "model.layers.{bid}.self_attn.v_proj",                       # llama-hf nemotron olmoe olmo2 phimoe
             "layers.{bid}.attention.wv",                                 # llama-pth
             "encoder.layer.{bid}.attention.self.value",                  # bert
             "transformer.h.{bid}.attn.v_proj",                           # gpt-j
@@ -190,7 +200,8 @@ class TensorNameMap:
             "transformer.blocks.{bid}.attn.out_proj",                       # mpt
             "transformer.h.{bid}.self_attention.dense",                     # falcon
             "h.{bid}.self_attention.dense",                                 # bloom
-            "model.layers.{bid}.self_attn.o_proj",                          # llama-hf nemotron olmoe olmo2
+            "model.layers.{bid}.self_attn.o_proj",                          # llama-hf nemotron olmoe olmo2 phimoe
+            "model.layers.{bid}.self_attn.linear_attn",                     # deci
             "layers.{bid}.attention.wo",                                    # llama-pth
             "encoder.layer.{bid}.attention.output.dense",                   # bert
             "transformer.h.{bid}.attn.out_proj",                            # gpt-j
@@ -234,7 +245,7 @@ class TensorNameMap:
             "transformer.h.{bid}.ln_2",                                      # gpt2 refact qwen jais exaone
             "h.{bid}.post_attention_layernorm",                              # bloom
             "transformer.blocks.{bid}.norm_2",                               # mpt
-            "model.layers.{bid}.post_attention_layernorm",                   # llama-hf nemotron olmoe
+            "model.layers.{bid}.post_attention_layernorm",                   # llama-hf nemotron olmoe phimoe
             "layers.{bid}.ffn_norm",                                         # llama-pth
             "language_model.encoder.layers.{bid}.post_attention_layernorm",  # persimmon
             "model.layers.{bid}.ln2",                                        # yi
@@ -257,7 +268,7 @@ class TensorNameMap:
 
         MODEL_TENSOR.FFN_GATE_INP: (
             "layers.{bid}.feed_forward.gate",                   # mixtral
-            "model.layers.{bid}.block_sparse_moe.gate",         # mixtral
+            "model.layers.{bid}.block_sparse_moe.gate",         # mixtral phimoe
             "model.layers.{bid}.mlp.gate",                      # qwen2moe olmoe
             "transformer.decoder_layer.{bid}.router",           # Grok
             "transformer.blocks.{bid}.ffn.router.layer",        # dbrx
@@ -268,6 +279,10 @@ class TensorNameMap:
             "model.layers.{bid}.mlp.shared_expert_gate", # qwen2moe
         ),
 
+        MODEL_TENSOR.FFN_EXP_PROBS_B: (
+            "model.layers.{bid}.mlp.gate.e_score_correction", # deepseek-v3
+        ),
+
         # Feed-forward up
         MODEL_TENSOR.FFN_UP: (
             "gpt_neox.layers.{bid}.mlp.dense_h_to_4h",                # gptneox
@@ -298,15 +313,16 @@ class TensorNameMap:
         ),
 
         MODEL_TENSOR.FFN_UP_EXP: (
-            "layers.{bid}.feed_forward.experts.w3",          # mixtral (merged)
-            "transformer.decoder_layer.{bid}.moe.linear_v",  # Grok (merged)
-            "transformer.blocks.{bid}.ffn.experts.mlp.v1",   # dbrx
-            "model.layers.{bid}.mlp.experts.up_proj",        # qwen2moe olmoe (merged)
+            "layers.{bid}.feed_forward.experts.w3",           # mixtral (merged)
+            "transformer.decoder_layer.{bid}.moe.linear_v",   # Grok (merged)
+            "transformer.blocks.{bid}.ffn.experts.mlp.v1",    # dbrx
+            "model.layers.{bid}.mlp.experts.up_proj",         # qwen2moe olmoe (merged)
+            "model.layers.{bid}.block_sparse_moe.experts.w3", # phimoe (merged)
         ),
 
         MODEL_TENSOR.FFN_UP_SHEXP: (
             "model.layers.{bid}.mlp.shared_expert.up_proj",  # qwen2moe
-            "model.layers.{bid}.mlp.shared_experts.up_proj", # deepseek2
+            "model.layers.{bid}.mlp.shared_experts.up_proj", # deepseek deepseek2
         ),
 
         # AWQ-activation gate
@@ -330,15 +346,16 @@ class TensorNameMap:
         ),
 
         MODEL_TENSOR.FFN_GATE_EXP: (
-            "layers.{bid}.feed_forward.experts.w1",         # mixtral (merged)
-            "transformer.decoder_layer.{bid}.moe.linear",   # Grok (merged)
-            "transformer.blocks.{bid}.ffn.experts.mlp.w1",  # dbrx
-            "model.layers.{bid}.mlp.experts.gate_proj",     # qwen2moe olmoe (merged)
+            "layers.{bid}.feed_forward.experts.w1",           # mixtral (merged)
+            "transformer.decoder_layer.{bid}.moe.linear",     # Grok (merged)
+            "transformer.blocks.{bid}.ffn.experts.mlp.w1",    # dbrx
+            "model.layers.{bid}.mlp.experts.gate_proj",       # qwen2moe olmoe (merged)
+            "model.layers.{bid}.block_sparse_moe.experts.w1", # phimoe (merged)
         ),
 
         MODEL_TENSOR.FFN_GATE_SHEXP: (
             "model.layers.{bid}.mlp.shared_expert.gate_proj",  # qwen2moe
-            "model.layers.{bid}.mlp.shared_experts.gate_proj", # deepseek2
+            "model.layers.{bid}.mlp.shared_experts.gate_proj", # deepseek deepseek2
         ),
 
         # Feed-forward down
@@ -375,11 +392,12 @@ class TensorNameMap:
             "transformer.blocks.{bid}.ffn.experts.mlp.w2",       # dbrx
             "model.layers.{bid}.mlp.experts.down_proj",          # qwen2moe olmoe (merged)
             "model.layers.{bid}.block_sparse_moe.output_linear", # granitemoe
+            "model.layers.{bid}.block_sparse_moe.experts.w2",    # phimoe (merged)
         ),
 
         MODEL_TENSOR.FFN_DOWN_SHEXP: (
             "model.layers.{bid}.mlp.shared_expert.down_proj",  # qwen2moe
-            "model.layers.{bid}.mlp.shared_experts.down_proj", # deepseek2
+            "model.layers.{bid}.mlp.shared_experts.down_proj", # deepseek deepseek2
         ),
 
         MODEL_TENSOR.ATTN_Q_NORM: (
@@ -449,34 +467,42 @@ class TensorNameMap:
 
         MODEL_TENSOR.TIME_MIX_W1: (
             "rwkv.blocks.{bid}.attention.time_maa_w1",  # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_w1", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_W2: (
             "rwkv.blocks.{bid}.attention.time_maa_w2",  # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_w2", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_X: (
             "rwkv.blocks.{bid}.attention.time_maa_x",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_x",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_K: (
             "rwkv.blocks.{bid}.attention.time_maa_k",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_k",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_V: (
             "rwkv.blocks.{bid}.attention.time_maa_v",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_v",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_R: (
             "rwkv.blocks.{bid}.attention.time_maa_r",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_r",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_G: (
             "rwkv.blocks.{bid}.attention.time_maa_g",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_g",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LERP_W: (
             "rwkv.blocks.{bid}.attention.time_maa_w",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_maa_w",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_FIRST: (
@@ -485,30 +511,37 @@ class TensorNameMap:
 
         MODEL_TENSOR.TIME_MIX_DECAY: (
             "rwkv.blocks.{bid}.attention.time_decay",   # rwkv v6
+            "model.layers.{bid}.self_attn.time_decay",  # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_DECAY_W1: (
             "rwkv.blocks.{bid}.attention.time_decay_w1",  # rwkv v6
+            "model.layers.{bid}.self_attn.time_decay_w1", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_DECAY_W2: (
             "rwkv.blocks.{bid}.attention.time_decay_w2",  # rwkv v6
+            "model.layers.{bid}.self_attn.time_decay_w2", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_KEY: (
-            "rwkv.blocks.{bid}.attention.key", # rwkv
+            "rwkv.blocks.{bid}.attention.key",     # rwkv
+            "model.layers.{bid}.self_attn.k_proj", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_VALUE: (
-            "rwkv.blocks.{bid}.attention.value", # rwkv
+            "rwkv.blocks.{bid}.attention.value",   # rwkv
+            "model.layers.{bid}.self_attn.v_proj", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_RECEPTANCE: (
             "rwkv.blocks.{bid}.attention.receptance", # rwkv
+            "model.layers.{bid}.self_attn.q_proj",    # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_GATE: (
-            "rwkv.blocks.{bid}.attention.gate", # rwkv
+            "rwkv.blocks.{bid}.attention.gate",  # rwkv
+            "model.layers.{bid}.self_attn.gate", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.TIME_MIX_LN: (
@@ -516,7 +549,8 @@ class TensorNameMap:
         ),
 
         MODEL_TENSOR.TIME_MIX_OUTPUT: (
-            "rwkv.blocks.{bid}.attention.output", # rwkv
+            "rwkv.blocks.{bid}.attention.output",  # rwkv
+            "model.layers.{bid}.self_attn.o_proj", # rwkv6qwen2
         ),
 
         MODEL_TENSOR.CHANNEL_MIX_LERP_K: (
@@ -681,6 +715,8 @@ class TensorNameMap:
             "encoder.block.{bid}.layer.1.DenseReluDense.wo", # t5
         ),
 
+        ############################################################################
+        # TODO: these do not belong to block_mappings_cfg - move them to mappings_cfg
         MODEL_TENSOR.ENC_OUTPUT_NORM: (
             "encoder.final_layer_norm", # t5
         ),
@@ -693,6 +729,67 @@ class TensorNameMap:
         MODEL_TENSOR.CLS_OUT: (
             "classifier.out_proj", # roberta
         ),
+        #############################################################################
+
+        MODEL_TENSOR.CONVNEXT_DW: (
+            "backbone.convnext.{bid}.dwconv", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.CONVNEXT_NORM: (
+            "backbone.convnext.{bid}.norm", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.CONVNEXT_PW1: (
+            "backbone.convnext.{bid}.pwconv1", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.CONVNEXT_PW2: (
+            "backbone.convnext.{bid}.pwconv2", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.CONVNEXT_GAMMA: (
+            "backbone.convnext.{bid}.gamma", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_CONV1: (
+            "backbone.posnet.{bid}.conv1", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_CONV2: (
+            "backbone.posnet.{bid}.conv2", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_NORM: (
+            "backbone.posnet.{bid}.norm", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_NORM1: (
+            "backbone.posnet.{bid}.norm1", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_NORM2: (
+            "backbone.posnet.{bid}.norm2", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_ATTN_NORM: (
+            "backbone.posnet.{bid}.norm", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_ATTN_Q: (
+            "backbone.posnet.{bid}.q", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_ATTN_K: (
+            "backbone.posnet.{bid}.k", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_ATTN_V: (
+            "backbone.posnet.{bid}.v", # wavtokenizer
+        ),
+
+        MODEL_TENSOR.POSNET_ATTN_OUT: (
+            "backbone.posnet.{bid}.proj_out", # wavtokenizer
+        ),
     }
 
     # architecture-specific block mappings

bigdl/cpp/gguf-py/gguf/utility.py

@@ -47,7 +47,7 @@ def size_label(total_params: int, shared_params: int, expert_params: int, expert
 
 
 def naming_convention(model_name: str | None, base_name: str | None, finetune_string: str | None, version_string: str | None, size_label: str | None, output_type: str | None, model_type: Literal['vocab', 'LoRA'] | None = None) -> str:
-    # Reference: https://github.com/ggerganov/ggml/blob/master/docs/gguf.md#gguf-naming-convention
+    # Reference: https://github.com/ggml-org/ggml/blob/master/docs/gguf.md#gguf-naming-convention
 
     if base_name is not None:
         name = base_name.strip().replace(' ', '-').replace('/', '-')

bigdl/cpp/gguf-py/gguf/vocab.py

@@ -127,7 +127,7 @@ class SpecialVocab:
                         self.merges = merges
                     elif isinstance(merges[0], list) and len(merges[0]) == 2 and isinstance(merges[0][0], str):
                         # New format since transformers 4.45 to support spaces in merges
-                        # ref: https://github.com/ggerganov/llama.cpp/issues/9692
+                        # ref: https://github.com/ggml-org/llama.cpp/issues/9692
                         # TODO: internally store as the new format instead of converting to old
                         if any(' ' in s for pair in merges for s in pair):
                             logger.warning(f'Spaces in merges detected, encoding as {chr(ord(" ") + 256)!r}')

{bigdl_core_cpp-2.6.0b20250320.dist-info → bigdl_core_cpp-2.6.0b20250321.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: bigdl-core-cpp
-Version: 2.6.0b20250320
+Version: 2.6.0b20250321
 Summary: Large Language Model Develop Toolkit
 Author: BigDL Authors
 License: Apache License, Version 2.0

bigdl_core_cpp-2.6.0b20250321.dist-info/RECORD

@@ -0,0 +1,57 @@
+bigdl/cpp/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+bigdl/cpp/convert_hf_to_gguf.py,sha256=GB6mGc_deGraPhQfUgU8i33odUb6WfMw0vVPcgZ_-ow,240529
+bigdl/cpp/convert_hf_to_gguf_update.py,sha256=1BFKEkj0BMDB90lUB5p_-iR9rSVcjgYPGWmEw28avB8,17721
+bigdl/cpp/convert_llama_ggml_to_gguf.py,sha256=0dKjRhmFzvWV4e-cuLmaeW14JrWUtZwerBmz8mYyMvI,19556
+bigdl/cpp/convert_lora_to_gguf.py,sha256=sHrcutdgzrDR5H7ZiLPOLoMnkJKg8uZ7OcFhAZhPrLo,19073
+bigdl/cpp/gguf-py/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+bigdl/cpp/gguf-py/gguf/__init__.py,sha256=h5GWs6SMXYR8giWZ7MTZzAc3hYsIJF-HAkdxtgXLOPo,228
+bigdl/cpp/gguf-py/gguf/constants.py,sha256=CJ0LigNqlnEqYP8IhnJsKcst9fIm-huE4RccvkTYUbg,69188
+bigdl/cpp/gguf-py/gguf/gguf.py,sha256=QpLc-xU055W2d7CEFvJp2gLIfGO63bdM24ZndZCH6rw,493
+bigdl/cpp/gguf-py/gguf/gguf_reader.py,sha256=PUrx08ZwaUOz1gLw5JQ459Hi7JIeCdlHgZX7wXcTqbI,12702
+bigdl/cpp/gguf-py/gguf/gguf_writer.py,sha256=pFgnwrsDupKxI3SHNQbfiuz7dUopCOqj3ERBPuZMkMo,39955
+bigdl/cpp/gguf-py/gguf/lazy.py,sha256=YIYxGBWD-oKXU4HOvpHs9eiEn81HUgeSmt1mmHJlbdM,8814
+bigdl/cpp/gguf-py/gguf/metadata.py,sha256=oBTb4DXi_h1L_gYm8x_JRVuEPR4GHlVHuM-iN0OxWoY,33244
+bigdl/cpp/gguf-py/gguf/py.typed,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
+bigdl/cpp/gguf-py/gguf/quants.py,sha256=2z6vcK-kBefqZbYNmSEVmdZF_tXHeVb5NC6jCbBdgKc,62040
+bigdl/cpp/gguf-py/gguf/tensor_mapping.py,sha256=w1JZkRIKHj2tKYADLDUvCOsQfVf8y7Y0ZwqmtmrtLBA,39549
+bigdl/cpp/gguf-py/gguf/utility.py,sha256=Mx4mqamXtatL15LCH04mG-7SNBwPzP2T75ts0uBnEuI,3002
+bigdl/cpp/gguf-py/gguf/vocab.py,sha256=QTdt4HZrn7onHqm2tpHTaEq2sL3yG07zbHqQR9iVfu8,20815
+bigdl/cpp/libs/common.lib,sha256=Vo5qx0PPK6ZyT0Z3TUoVqPhxI-A-zZR7IHWSeLYTIkM,8642856
+bigdl/cpp/libs/ggml-base.dll,sha256=IUkUVt54mNJzwdTXmwvyJ_nQLOqmWrUpIrpRCg7VPFE,577024
+bigdl/cpp/libs/ggml-cpu.dll,sha256=BTX-iKt3_8YgJcpRl0pXDrES2ZYTEy9ph6i4mMYwLqQ,1038336
+bigdl/cpp/libs/ggml-sycl.dll,sha256=aqQdnqwEY-JlKm4Dny8nhy3OgSySYaiDWGSXwIp2SHc,5560832
+bigdl/cpp/libs/ggml.dll,sha256=XjdK5jlGNhAxr7_IGiTWkvi4Z06ciDqiL5wbW9ZRuVc,118272
+bigdl/cpp/libs/libc++.dll,sha256=U0TVK2WfFQIJPP6Bz9SeJmgskm2iqZWJorx_DGdfKIw,1561600
+bigdl/cpp/libs/llama-batched.exe,sha256=v27oJ9gdolpQArt7Ih8m3qXCrIcEhWmgJrbVoxwpy0Q,1741824
+bigdl/cpp/libs/llama-bench.exe,sha256=4aZtBZ2Bs-Q5kXVTd86WrE1uF3LuJCWfGaJcfHTkAH8,279552
+bigdl/cpp/libs/llama-cli.exe,sha256=Sa6UWO5May8Ub1tgRiLI3gBJ1ayJbiZwQtshD9Ckdu0,1812480
+bigdl/cpp/libs/llama-embedding.exe,sha256=WGqbovh_2njmpN1OE-5FU5yH-6D8hRQQ_cDKj_GlXlQ,1765376
+bigdl/cpp/libs/llama-gemma3-cli.exe,sha256=94z47LKR7CrVSZ2w_tzuHJ66hJhw7v6BWrvbZIS1r_U,2033664
+bigdl/cpp/libs/llama-gguf.exe,sha256=DlBsPv7pL0Rsp5hafby51gxRnXmzsti_569wySHAQ-Q,59392
+bigdl/cpp/libs/llama-llava-cli.exe,sha256=DyoTXM5HOGMQANjfZbsYk50EsxZK5sBGVDuTfYX6Adg,2019840
+bigdl/cpp/libs/llama-lookup.exe,sha256=lievf9j2yNC5uujL0oMmrpBj7J14PZWwi2XT_DX6_3U,1801216
+bigdl/cpp/libs/llama-ls-sycl-device.exe,sha256=ZNfBEekcD4AiaNPkqafa1pzfo8WTp0DpS7I6Mw-RAwo,10240
+bigdl/cpp/libs/llama-minicpmv-cli.exe,sha256=sozeVIjDxfa4NssDmTPBOO-6vgK1nJ-Bs5HDQp7QBE8,2017792
+bigdl/cpp/libs/llama-perplexity.exe,sha256=vPzXHFomoFPgtLznM1fw1JQKDv_4q4bSqUuBb1qKEpI,1886208
+bigdl/cpp/libs/llama-quantize.exe,sha256=CV3wKKGunBiBy6E8fStqXeFefjAnBmJUmXGwbMPyPb0,122880
+bigdl/cpp/libs/llama-server.exe,sha256=haDGFy0qNNR_WUys-9CaDTB3RJwjvrBkCKBlhuewRBU,4130304
+bigdl/cpp/libs/llama-simple.exe,sha256=Wfq0JYyVA4itZXPKs20UAOs0z-Q59SWOwpSlRgUWUw0,62464
+bigdl/cpp/libs/llama-speculative.exe,sha256=rQG1WQizxNX0oj1nzvxxn0k3DCkkOkk0flLgw0_B2fs,1803776
+bigdl/cpp/libs/llama-tokenize.exe,sha256=LA0ZLlq2Akt_CXTz8lSXw6-S9aCUnBm8RhEubpVZaz8,89088
+bigdl/cpp/libs/llama.dll,sha256=W07-VmnRG1Hm7NkUdA9lFjrW9Wz81pQlKfZWF8yugRI,1470464
+bigdl/cpp/libs/llava_shared.dll,sha256=IY0l5XnYnL014k398AHe33wTqwuBhKc1wTgn_uMeciA,380416
+bigdl/cpp/libs/ollama-ggml-base.dll,sha256=ma9wJKTWKvr16EAlGJPwuNJN-dwUCx-gYPAacF0HO5U,459776
+bigdl/cpp/libs/ollama-ggml-cpu.dll,sha256=lW9Ssioxrjm60ii9-LiLBoYJiFCpva8naloJpeGrjnI,477184
+bigdl/cpp/libs/ollama-ggml-sycl.dll,sha256=I-Ksd6w8OGivjKsPwdNA0u7sGzPvyFwu1W5OvJi2IcA,5326336
+bigdl/cpp/libs/ollama-lib.exe,sha256=LAaaV9voEPaeGBSSLsF0Eb9T0_zibH4SIe7NBakYses,25916416
+bigdl/cpp/libs/ollama.exe,sha256=GceJtJEhtrlNDAY9n-7FMyWiLoHP6UqgwOrrM7RdUIk,207360
+bigdl/cpp/libs/ollama_ggml.dll,sha256=X8XsTQd6Uc-LgMOzjNvGpnemqevACWoTVYdx7ZK6Zbc,113152
+bigdl/cpp/libs/ollama_llama.dll,sha256=XKb_ypeqhlL9_lP_AcxNR6Z8xdQcr-3Inr6l-CzX1o4,1421312
+bigdl/cpp/libs/ollama_llava_shared.dll,sha256=LtLN0l-jXVF8BNSH8XZAaAuIZXQWrc9cssBlPnYCixQ,398336
+bigdl_core_cpp-2.6.0b20250321.data/scripts/init-llama-cpp.bat,sha256=U0h6RifZxL3GGJp-0dxdZapQIvXUATSj644CURJL-lg,751
+bigdl_core_cpp-2.6.0b20250321.data/scripts/init-llama-cpp.ps1,sha256=JFOylLxO4MKpllHhdbPuJ1xHi9azxDpzdJns8JtZpkU,501
+bigdl_core_cpp-2.6.0b20250321.data/scripts/init-ollama.bat,sha256=0I2iBOFv3kR9hvEySGMRUU52-qwVhE7oRZnyWz-2z_U,657
+bigdl_core_cpp-2.6.0b20250321.dist-info/METADATA,sha256=9I4TcjPb4JO3gGlDHDAXgFbUhCRqhuZp9iOyVFX2Apo,750
+bigdl_core_cpp-2.6.0b20250321.dist-info/WHEEL,sha256=pUQ3YzM9z7CMLK4Pdg7RxRLrm1NUy0aQs4ESywX3iFk,97
+bigdl_core_cpp-2.6.0b20250321.dist-info/top_level.txt,sha256=iGuLfZARD_qANcIMfy0tbbrC3EtCg6BSiH8icc3dLWs,6
+bigdl_core_cpp-2.6.0b20250321.dist-info/RECORD,,

{bigdl_core_cpp-2.6.0b20250320.dist-info → bigdl_core_cpp-2.6.0b20250321.dist-info}/WHEEL

@@ -1,5 +1,5 @@
 Wheel-Version: 1.0
-Generator: setuptools (77.0.1)
+Generator: setuptools (77.0.3)
 Root-Is-Purelib: true
 Tag: py3-none-win_amd64