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-hparams.cpp

@@ -139,6 +139,13 @@ uint32_t llama_hparams::n_embd_r() const {
         return n_embd * (n_shortconv_l_cache - 1);
     }
 
+    if (n_embd_head_kda != 0) {
+        // for Kimi KDA layers
+        // Conv state for Q, K, V: 3 * (d_conv - 1) * n_head * head_dim
+        const uint32_t d_inner = n_head() * n_embd_head_kda;  // 32 * 128 = 4096
+        return 3 * (ssm_d_conv > 0 ? ssm_d_conv - 1 : 3) * d_inner;
+    }
+
     // TODO: maybe support other convolution strides than 1
     // NOTE: since the first column of the conv_state is shifted out each time, it's not actually needed
     // Corresponds to Mamba's conv_states size
@@ -151,6 +158,13 @@ uint32_t llama_hparams::n_embd_s() const {
         return n_embd * wkv_head_size;
     }
 
+    if (n_embd_head_kda != 0) {
+        // for Kimi KDA layers
+        // Full recurrent state: head_dim * head_dim * n_head
+        // h tensor shape for delta attention: [head_dim, head_dim, n_head]
+        return n_embd_head_kda * n_embd_head_kda * n_head();  // 128 * 128 * 32 = 524288
+    }
+
     // corresponds to Mamba's ssm_states size
     return ssm_d_state * ssm_d_inner;
 }