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* rm outdated test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-model.cpp

@@ -11,6 +11,7 @@
 #include "llama-kv-cache.h"
 #include "llama-kv-cache-iswa.h"
 #include "llama-kv-cache-dsa.h"
+#include "llama-kv-cache-dsv4.h"
 #include "llama-memory-hybrid.h"
 #include "llama-memory-hybrid-iswa.h"
 #include "llama-memory-recurrent.h"
@@ -181,6 +182,8 @@ static llama_model * llama_model_mapping(llm_arch arch, const llama_model_params
             return new llama_model_deepseek2ocr(params);
         case LLM_ARCH_DEEPSEEK32:
             return new llama_model_deepseek32(params);
+        case LLM_ARCH_DEEPSEEK4:
+            return new llama_model_deepseek4(params);
         case LLM_ARCH_GLM_DSA:
             return new llama_model_glm_dsa(params);
         case LLM_ARCH_MISTRAL4:
@@ -817,6 +820,7 @@ static const char * llama_expert_gating_func_name(llama_expert_gating_func_type
     switch (type) {
         case LLAMA_EXPERT_GATING_FUNC_TYPE_SOFTMAX: return "softmax";
         case LLAMA_EXPERT_GATING_FUNC_TYPE_SIGMOID: return "sigmoid";
+        case LLAMA_EXPERT_GATING_FUNC_TYPE_SQRT_SOFTPLUS: return "sqrtsoftplus";
         default:                                    return "unknown";
     }
 }
@@ -2156,7 +2160,24 @@ llama_memory_i * llama_model::create_memory(const llama_memory_params & params,
                         }
                     }
 
-                    if (hparams.swa_type != LLAMA_SWA_TYPE_NONE) {
+                    if (arch == LLM_ARCH_DEEPSEEK4) {
+                        GGML_ASSERT(hparams.swa_type != LLAMA_SWA_TYPE_NONE);
+
+                        res = new llama_kv_cache_dsv4(
+                                *this,
+                                params.type_k,
+                                params.type_v,
+                                !cparams.flash_attn,
+                                cparams.offload_kqv,
+                                params.swa_full,
+                                cparams.kv_unified,
+                                cparams.n_ctx_seq,
+                                cparams.n_seq_max,
+                                cparams.n_ubatch,
+                                1,
+                                filter,
+                                reuse);
+                    } else if (hparams.swa_type != LLAMA_SWA_TYPE_NONE) {
                         GGML_ASSERT(hparams.is_swa_any());
 
                         if (arch == LLM_ARCH_GEMMA4_ASSISTANT) {
@@ -2328,6 +2349,11 @@ int32_t llama_model_n_head_kv(const llama_model * model) {
 }
 
 int32_t llama_model_n_swa(const llama_model * model) {
+    // dsv4 kv-cache has SWA but it cannot be used as a rollback because of
+    // other compression ratios, so we return 0 here
+    if (model->arch == LLM_ARCH_DEEPSEEK4) {
+        return 0;
+    }
     return model->hparams.n_swa;
 }
 
@@ -2409,6 +2435,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
         case LLM_ARCH_DEEPSEEK2:
         case LLM_ARCH_DEEPSEEK2OCR:
         case LLM_ARCH_DEEPSEEK32:
+        case LLM_ARCH_DEEPSEEK4:
         case LLM_ARCH_PLM:
         case LLM_ARCH_CHATGLM:
         case LLM_ARCH_GRANITE: