Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

model : support for DeepseekV32ForCausalLM with generic DeepSeek Sparse Attention (DSA) implementation (#23346)

Browse Source 
* llama : support DeepSeek V3.2 model family (with DSA lightning indexer)

* convert : handle DeepseekV32ForCausalLM architecture

* ggml : support for f16 GGML_OP_FILL

* memory : separate hparams argument in llama_kv_cache constructor

* memory : add llama_kv_cache_dsa memory (KV cache + lightning indexer cache)

* llama : support for LLM_ARCH_DEEPSEEK32

* model : llama_model_deepseek32 implementation

* model : merge two scale operations into one in DSA lightning indexer implementation

* chore : remove unused code

* model : support NVFP4 in DeepSeek V3.2

Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>

* memory : refactoring TODO

Co-authored-by: ggerganov <ggerganov@users.noreply.github.com>

---------

Co-authored-by: Stanisław Szymczyk <sszymczy@gmail.com>
Co-authored-by: Sigbjørn Skjæret <sigbjorn.skjaeret@scala.com>
Co-authored-by: ggerganov <ggerganov@users.noreply.github.com>
This commit is contained in:
fairydreaming
2026-05-29 10:15:17 +02:00
committed by GitHub
co-authored by Stanisław Szymczyk Sigbjørn Skjæret ggerganov
parent 031ddb2e08
commit 1f0aa2a696
22 changed files with 1261 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/llama-model.cpp
+24 -1
View File
@@ -10,6 +10,7 @@
#include "llama-kv-cache.h"
#include "llama-kv-cache-iswa.h"
#include "llama-kv-cache-dsa.h"
#include "llama-memory-hybrid.h"
#include "llama-memory-hybrid-iswa.h"
#include "llama-memory-recurrent.h"
@@ -172,6 +173,8 @@ static llama_model * llama_model_mapping(llm_arch arch, const llama_model_params
return new llama_model_deepseek2(params);
case LLM_ARCH_DEEPSEEK2OCR:
return new llama_model_deepseek2ocr(params);
case LLM_ARCH_DEEPSEEK32:
return new llama_model_deepseek32(params);
case LLM_ARCH_GLM_DSA:
return new llama_model_glm_dsa(params);
case LLM_ARCH_MISTRAL4:
@@ -779,6 +782,7 @@ const char * llm_type_name(llm_type type) {
case LLM_TYPE_310B_A15B: return "310B.A15B";
case LLM_TYPE_355B_A32B: return "355B.A32B";
case LLM_TYPE_397B_A17B: return "397B.A17B";
case LLM_TYPE_685B_A37B: return "685B.A37B";
case LLM_TYPE_744B_A40B: return "744B.A40B";
case LLM_TYPE_E2B: return "E2B";
case LLM_TYPE_E4B: return "E4B";
@@ -1769,7 +1773,7 @@ void llama_model::print_info() const {
LLAMA_LOG_INFO("%s: expert_weights_scale = %.1f\n", __func__, hparams.expert_weights_scale);
}
if (arch == LLM_ARCH_DEEPSEEK2 || arch == LLM_ARCH_DEEPSEEK2OCR || arch == LLM_ARCH_GLM_DSA || arch == LLM_ARCH_MISTRAL4) {
if (arch == LLM_ARCH_DEEPSEEK2 || arch == LLM_ARCH_DEEPSEEK2OCR || arch == LLM_ARCH_DEEPSEEK32 || arch == LLM_ARCH_GLM_DSA || arch == LLM_ARCH_MISTRAL4) {
LLAMA_LOG_INFO("%s: n_layer_dense_lead = %d\n", __func__, hparams.n_layer_dense_lead);
LLAMA_LOG_INFO("%s: n_lora_q = %d\n", __func__, hparams.n_lora_q);
LLAMA_LOG_INFO("%s: n_lora_kv = %d\n", __func__, hparams.n_lora_kv);
@@ -1957,6 +1961,23 @@ llama_memory_i * llama_model::create_memory(const llama_memory_params & params,
{
res = nullptr;
} break;
case LLM_ARCH_DEEPSEEK32:
{
res = new llama_kv_cache_dsa(
*this,
params.type_k,
params.type_v,
!cparams.flash_attn,
cparams.offload_kqv,
cparams.kv_unified,
cparams.n_ctx_seq,
cparams.n_seq_max,
1,
hparams.n_swa,
hparams.swa_type,
nullptr,
nullptr);
} break;
// Models that need standard caching should rely on recurrent/hybrid
// checks
default:
@@ -2083,6 +2104,7 @@ llama_memory_i * llama_model::create_memory(const llama_memory_params & params,
res = new llama_kv_cache(
*this,
hparams,
params.type_k,
params.type_v,
!cparams.flash_attn,
@@ -2272,6 +2294,7 @@ llama_rope_type llama_model_rope_type(const llama_model * model) {
case LLM_ARCH_DEEPSEEK:
case LLM_ARCH_DEEPSEEK2:
case LLM_ARCH_DEEPSEEK2OCR:
case LLM_ARCH_DEEPSEEK32:
case LLM_ARCH_PLM:
case LLM_ARCH_CHATGLM:
case LLM_ARCH_GRANITE: