DeepSeek V4 (#24162)

* convert: add dsv4 conversion

* add basic setup

* add llm_graph_input_dsv4

* add save-load state

* add sinkhorn eps - correction by @fairydreaming

* add rope fix

* cleanup dead code

* fix bugs

* support pro model: added by @fairydreaming

* remove redundant V cache

* Chat template

* remove debugging leftovers

* Add mechanism for inlining templates based on architecture

* s/deepseek-v4-flash/deepseek4/g

* s/deepseek-v4-flash/deepseek4/g continued

* enable graph reuse

* enable FA

* fix test llama archs

* rename

* compatibility with antirez ds4 GGUFs

* simplified set_gguf_parameters() by calling super class method, replaced moe.score_func with expert_gating_func.

* reserve worst-case kv-cache

* revert max split inputs

* address review comments

* add padding to enable FA

* pad only the final value of plan.n_kv to 256

* remove built-in cpp chat template

* cont: remove cpp built-in template

* rm outdated test

* replace ggml_view_3d() with ggml_reshape_3d()

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

* only support n_seq=1 for now

* remove unused var

* cont: remove unused var

* use scale bias

* use correct ptr for can_reuse

* remove gen-chat-inline-templates.py

* simplify graph reuse

* cont: cleanup

* remove unused inputs

* enable partial checkpointing

* add correct shape for kq_mask + set llama_model_n_swa to 0 for dsv4

* precompute source_idx + add comment about dummy write

* support multi-seq

* remove restored_trim_pos

* use split_equal when possible

* fix indent

* address review comments

* use LLM_KV

* fix ci

---------

Co-authored-by: Piotr Wilkin <piotr.wilkin@syndatis.com>
Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Co-authored-by: Xuan Son Nguyen <son@huggingface.co>
Co-authored-by: fairydreaming <166155368+fairydreaming@users.noreply.github.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

src/llama-arch.h

@@ -82,6 +82,7 @@ enum llm_arch {
     LLM_ARCH_DEEPSEEK2,
     LLM_ARCH_DEEPSEEK2OCR,
     LLM_ARCH_DEEPSEEK32,
+    LLM_ARCH_DEEPSEEK4,
     LLM_ARCH_CHATGLM,
     LLM_ARCH_GLM4,
     LLM_ARCH_GLM4_MOE,
@@ -255,9 +256,19 @@ enum llm_kv {
     LLM_KV_ATTENTION_INDEXER_HEAD_COUNT,
     LLM_KV_ATTENTION_INDEXER_KEY_LENGTH,
     LLM_KV_ATTENTION_INDEXER_TOP_K,
+    LLM_KV_ATTENTION_OUTPUT_GROUP_COUNT,
+    LLM_KV_ATTENTION_OUTPUT_LORA_RANK,
+    LLM_KV_ATTENTION_COMPRESS_ROPE_FREQ_BASE,
+    LLM_KV_ATTENTION_COMPRESS_RATIOS,
     LLM_KV_ATTENTION_SHARED_KV_LAYERS,
     LLM_KV_ATTENTION_RECURRENT_LAYERS,
 
+    LLM_KV_HYPER_CONNECTION_COUNT,
+    LLM_KV_HYPER_CONNECTION_SINKHORN_ITERATIONS,
+    LLM_KV_HYPER_CONNECTION_EPSILON,
+
+    LLM_KV_HASH_LAYER_COUNT,
+
     LLM_KV_ROPE_DIMENSION_COUNT,
     LLM_KV_ROPE_DIMENSION_COUNT_SWA,
     LLM_KV_ROPE_DIMENSION_SECTIONS,
@@ -501,10 +512,27 @@ enum llm_tensor {
     LLM_TENSOR_ATTN_Q_B,
     LLM_TENSOR_ATTN_KV_A_MQA,
     LLM_TENSOR_ATTN_KV_B,
+    LLM_TENSOR_ATTN_KV,
+    LLM_TENSOR_ATTN_KV_NORM,
+    LLM_TENSOR_ATTN_OUT_A,
+    LLM_TENSOR_ATTN_OUT_B,
     LLM_TENSOR_ATTN_K_B,
     LLM_TENSOR_ATTN_V_B,
     LLM_TENSOR_ATTN_Q_A_NORM,
     LLM_TENSOR_ATTN_KV_A_NORM,
+    LLM_TENSOR_HC_HEAD_FN,
+    LLM_TENSOR_HC_HEAD_BASE,
+    LLM_TENSOR_HC_HEAD_SCALE,
+    LLM_TENSOR_HC_ATTN_FN,
+    LLM_TENSOR_HC_ATTN_BASE,
+    LLM_TENSOR_HC_ATTN_SCALE,
+    LLM_TENSOR_HC_FFN_FN,
+    LLM_TENSOR_HC_FFN_BASE,
+    LLM_TENSOR_HC_FFN_SCALE,
+    LLM_TENSOR_ATTN_COMPRESSOR_WKV,
+    LLM_TENSOR_ATTN_COMPRESSOR_WGATE,
+    LLM_TENSOR_ATTN_COMPRESSOR_APE,
+    LLM_TENSOR_ATTN_COMPRESSOR_NORM,
     LLM_TENSOR_ATTN_SUB_NORM,
     LLM_TENSOR_FFN_SUB_NORM,
     LLM_TENSOR_DEC_ATTN_NORM,
@@ -566,6 +594,11 @@ enum llm_tensor {
     LLM_TENSOR_INDEXER_PROJ,
     LLM_TENSOR_INDEXER_ATTN_K,
     LLM_TENSOR_INDEXER_ATTN_Q_B,
+    LLM_TENSOR_INDEXER_COMPRESSOR_WKV,
+    LLM_TENSOR_INDEXER_COMPRESSOR_WGATE,
+    LLM_TENSOR_INDEXER_COMPRESSOR_APE,
+    LLM_TENSOR_INDEXER_COMPRESSOR_NORM,
+    LLM_TENSOR_FFN_GATE_TID2EID,
     LLM_TENSOR_NEXTN_PROJ_PRE,
     LLM_TENSOR_NEXTN_PROJ_POST,
     LLM_TENSOR_NEXTN_EH_PROJ,