Kimi-Linear support (backend agnostic + MLA KV cache) (#18755)

* kimi linear model implementation

* kimi linear convert_hf_to_gguf

* kimi linear constants.py tensor_mapping.py

* Kimi Linear ggml.h

* kimi linear ggml-cpu

* Kimi Linear ggml-cuda

* Kimi Linear ggml.c

* kimi linear src/llama

* remove "const int64_t n_seq_tokens = q->ne[2];" to get rid of unused variable warning

* remove type mismatch warning

* read MoE params

* removed some hard coded code

* removed all hard code

* use DeepseekV2 tokenizer

* removed unnecessary internal methods called by the old set_vocab of KimiLinear

* rewrite get_vocab for KimiLinear. Removed all kda_scan code

* removed all traces of kda_scan

* reduce OP count by 1 due to removal of kda_scan

* Move KIMI_LINEAR to llm_arch_is_hybrid to enable KV cache

* set n_embd_head_k/v to ensure kv cache works

* don't quantize conv1d of Kimi Linear

* Kimi Linear backend agnostic

* removed LOG_INFO

* naive chunking form implemented

* fixed some comments

* add Kimi-K2 specific tokens to be recognized as EOG

* build_kda_autoregressive is implemented to replace build_kda_recurrent for faster inference. sync'd to b7682

* replaced Akk and Aqk with mul_mat and clamp

* no clamp version

* Moved Aqk computation out of the loop

* fixed typo and split wkv_b into wk_b and wv_b

* MLA KV cache support

* fix trailing spaces

* moved const llama_model & model; around to follow qwen3next format and see if it cna pass the -Wunused-private-field error

* fix trailing whitespace

* removed traling whitespaces in empty line + make sure indentation is multiple of 4

* try to make lint happy

* remove blank lines to make lint happy

* removed at least blank line containing white space

* fixed flake8 complaints locally

* return ggml_tensor * pair in kda_autoregressive and kda_chunking as in ngxson's Qwen3Next improvement

* removed Kimi-Linear specific change that causes failure at server-windows

* removed private: from kimi_linear to make build checks happy

* removed unnecessary ggml_cont before ggml_reshape

* created static function causal_conv1d to abtract similar code for q/k/v

* merged dt_bias to SSM_DT. Do -exp(log_A) in convert_hf_to_gguf.py.

* reverted to original

* fixed find_hparam calls. Fixed e_score_correction_bias to use bias instead of weight. Removed all ssm_conv bias terms.

* remove DT_B from constants.py. remove one comment line in llama-model.cpp

* new class llm_graph_input_mem_hybrid_k to get around the new MLA change. switch the concat order of ggml_concat calls in kimi-linear.cpp to accommodate MLA changes. Removed support for exp_probs_b.weight

* remove ssm_o_norm_b

* remove ssm_o_norm_b

* changed hparams.kda_head_dim to hparams.n_embd_head_kda. added TODO comment for class llama_graph_mem_hybrid_k

* removed all ggml_cont b4 ggml_reshape_4d

* Whitespace

* replaced all hparams.get with find_hparams

* added new names for n_experts, n_experts_used and score_func in TextModel and removed their code in KimiLinear in convert_hf_to_gguf.py. Removed unnecessary ggml_cont and GGML_ASSERT in kimi-linear.cpp

* use is_mla to switch between different mem_hybrid types

* fixed logical errors in convert_hf_to_gguf.py pointed out by CISC

* removed if else for required parameters kv_lora_rank and qk_rope_head_dim

* add back ggml_cont for Vcur

* minor changes

* removed extra line in llama-vocab.cpp. Added back the comment in llama-graph.cpp

* f16 gguf cannot run without context length

* made a mistake of adding back n_ctx parsing

---------

Co-authored-by: Piotr Wilkin (ilintar) <piotr.wilkin@syndatis.com>

src/llama-graph.cpp

@@ -533,6 +533,50 @@ bool llm_graph_input_mem_hybrid::can_reuse(const llm_graph_params & params) {
     return res;
 }
 
+// TODO: Hybrid input classes are a bit redundant.
+// Instead of creating a hybrid input, the graph can simply create 2 separate inputs.
+// Refactoring is required in the future.
+void llm_graph_input_mem_hybrid_k::set_input(const llama_ubatch * ubatch) {
+    mctx->get_attn()->set_input_k_idxs(inp_attn->self_k_idxs, ubatch);
+
+    mctx->get_attn()->set_input_kq_mask(inp_attn->self_kq_mask, ubatch, cparams.causal_attn);
+
+    const int64_t n_rs = mctx->get_recr()->get_n_rs();
+
+    if (inp_rs->s_copy) {
+        GGML_ASSERT(ggml_backend_buffer_is_host(inp_rs->s_copy->buffer));
+        int32_t * data = (int32_t *) inp_rs->s_copy->data;
+
+        // assuming copy destinations ALWAYS happen ONLY on the cells between head and head+n
+        for (uint32_t i = 0; i < n_rs; ++i) {
+            data[i] = mctx->get_recr()->s_copy(i);
+        }
+    }
+}
+
+bool llm_graph_input_mem_hybrid_k::can_reuse(const llm_graph_params & params) {
+    const auto * mctx = static_cast<const llama_memory_hybrid_context *>(params.mctx);
+
+    this->mctx = mctx;
+
+    bool res = true;
+
+    res &= inp_attn->self_k_idxs->ne[0] == params.ubatch.n_tokens;
+
+    res &= inp_attn->self_kq_mask->ne[0] == mctx->get_attn()->get_n_kv();
+    res &= inp_attn->self_kq_mask->ne[1] == params.ubatch.n_tokens;
+
+    res &= inp_rs->s_copy->ne[0] == mctx->get_recr()->get_n_rs();
+
+    res &= inp_rs->s_copy_main->ne[0]  == params.ubatch.n_seqs;
+    res &= inp_rs->s_copy_extra->ne[0] == mctx->get_recr()->get_n_rs() - params.ubatch.n_seqs;
+
+    res &= inp_rs->head == mctx->get_recr()->get_head();
+    res &= inp_rs->rs_z == mctx->get_recr()->get_rs_z();
+
+    return res;
+}
+
 void llm_graph_input_mem_hybrid_iswa::set_input(const llama_ubatch * ubatch) {
     const auto * attn_ctx = mctx->get_attn();
 
@@ -2268,6 +2312,17 @@ llm_graph_input_mem_hybrid * llm_graph_context::build_inp_mem_hybrid() const {
     return (llm_graph_input_mem_hybrid *) res->add_input(std::move(inp));
 }
 
+llm_graph_input_mem_hybrid_k * llm_graph_context::build_inp_mem_hybrid_k() const {
+    const auto * mctx_cur = static_cast<const llama_memory_hybrid_context *>(mctx);
+
+    auto inp_rs   = build_rs_inp_impl     (ctx0, ubatch, mctx_cur->get_recr());
+    auto inp_attn = build_attn_inp_k_impl(ctx0, ubatch, hparams, cparams, mctx_cur->get_attn());
+
+    auto inp = std::make_unique<llm_graph_input_mem_hybrid_k>(cparams, std::move(inp_attn), std::move(inp_rs), mctx_cur);
+
+    return (llm_graph_input_mem_hybrid_k *) res->add_input(std::move(inp));
+}
+
 llm_graph_input_mem_hybrid_iswa * llm_graph_context::build_inp_mem_hybrid_iswa() const {
     const auto * mctx_cur = static_cast<const llama_memory_hybrid_iswa_context *>(mctx);