model: move load_hparams and load_tensors to per-model definition (#22004)

* git-friendly migration * add build_graph * nits * exclude old code from build * wip * add llm_arch_model_i * prepare downstream functions * nits * nits * wip * wip * add back create_tensor_qkv * fix files missing include * enforce one llm_build per arch * cmake: use glob * missing model params * nits * wip * wip (2) * wip (3) * test-llama-archs is happy * improve switch case * move more stuff into llm_arch_model_i * fix downstream code * nits * nits (2) * fix order * llama_model_base * LLAMA_LOAD_LOCALS * small fix * fix build errors * auto * rm migration script and ifdef
2026-05-04 12:36:59 +02:00
parent c84e6d6db5
commit 994118a183
129 changed files with 10667 additions and 8117 deletions
@@ -1,6 +1,78 @@
 #include "models.h"

-llm_build_glm4::llm_build_glm4(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) {
+void llama_model_glm4::load_arch_hparams(llama_model_loader & ml) {
+    ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS,    hparams.f_norm_rms_eps);
+    ml.get_key_or_arr(LLM_KV_ROPE_DIMENSION_SECTIONS, hparams.rope_sections, 4, false);
+
+    // NextN/MTP parameters (GLM-OCR)
+    ml.get_key(LLM_KV_NEXTN_PREDICT_LAYERS, hparams.nextn_predict_layers, false);
+    GGML_ASSERT(hparams.nextn_predict_layers < hparams.n_layer && "nextn_predict_layers must be < n_layer");
+
+    // TODO: when MTP is implemented, this should probably be updated if needed
+    hparams.n_layer_kv_from_start = hparams.n_layer - hparams.nextn_predict_layers;
+
+    switch (hparams.n_layer) {
+        case 17: type = LLM_TYPE_1B; break; // GLM-OCR
+        case 40: type = LLM_TYPE_9B; break;
+        case 61: type = LLM_TYPE_32B; break;
+        default: type = LLM_TYPE_UNKNOWN;
+    }
+}
+
+void llama_model_glm4::load_arch_tensors(llama_model_loader &) {
+    LLAMA_LOAD_LOCALS;
+
+    tok_embd = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, 0);
+
+    // output
+    output_norm = create_tensor(tn(LLM_TENSOR_OUTPUT_NORM, "weight"), {n_embd}, 0);
+    output      = create_tensor(tn(LLM_TENSOR_OUTPUT,      "weight"), {n_embd, n_vocab}, TENSOR_NOT_REQUIRED);
+    // if output is NULL, init from the input tok embed
+    if (output == NULL) {
+        output = create_tensor(tn(LLM_TENSOR_TOKEN_EMBD, "weight"), {n_embd, n_vocab}, TENSOR_DUPLICATED);
+    }
+
+    for (int i = 0; i < n_layer; ++i) {
+        int flags = 0;
+        if (hparams.nextn_predict_layers > 0 && static_cast<uint32_t>(i) >= n_layer - hparams.nextn_predict_layers) {
+            // skip all tensors in the NextN layers
+            flags |= TENSOR_SKIP;
+        }
+
+        auto & layer = layers[i];
+
+        layer.attn_norm = create_tensor(tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, flags);
+        create_tensor_qkv(layer, i, n_embd, n_embd_head_k * n_head, n_embd_k_gqa, n_embd_v_gqa, flags);
+
+        layer.wo   = create_tensor(tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd_head_k * n_head, n_embd}, flags);
+
+        layer.attn_post_norm = create_tensor(tn(LLM_TENSOR_ATTN_POST_NORM, "weight", i), {n_embd}, flags);
+
+        layer.ffn_norm = create_tensor(tn(LLM_TENSOR_FFN_NORM, "weight", i), {n_embd}, flags);
+        layer.ffn_down = create_tensor(tn(LLM_TENSOR_FFN_DOWN, "weight", i), {  n_ff, n_embd}, flags);
+        layer.ffn_up   = create_tensor(tn(LLM_TENSOR_FFN_UP,   "weight", i), {n_embd, n_ff * 2}, flags);
+
+        layer.ffn_post_norm  = create_tensor(tn(LLM_TENSOR_FFN_POST_NORM, "weight", i), {n_embd}, flags);
+
+        // NextN/MTP tensors (preserved but unused) - conditionally load for last nextn_predict_layers
+        if (hparams.nextn_predict_layers > 0 && static_cast<uint32_t>(i) >= n_layer - hparams.nextn_predict_layers) {
+            layer.nextn.eh_proj          = create_tensor(tn(LLM_TENSOR_NEXTN_EH_PROJ, "weight", i), { 2 * n_embd, n_embd }, flags);
+            layer.nextn.enorm            = create_tensor(tn(LLM_TENSOR_NEXTN_ENORM, "weight", i), { n_embd }, flags);
+            layer.nextn.hnorm            = create_tensor(tn(LLM_TENSOR_NEXTN_HNORM, "weight", i), { n_embd }, flags);
+
+            // Optional tensors
+            layer.nextn.embed_tokens     = create_tensor(tn(LLM_TENSOR_NEXTN_EMBED_TOKENS, "weight", i), { n_embd, n_vocab }, flags | TENSOR_NOT_REQUIRED);
+            layer.nextn.shared_head_head = create_tensor(tn(LLM_TENSOR_NEXTN_SHARED_HEAD_HEAD, "weight", i), { n_embd, n_vocab }, flags | TENSOR_NOT_REQUIRED);
+            layer.nextn.shared_head_norm = create_tensor(tn(LLM_TENSOR_NEXTN_SHARED_HEAD_NORM, "weight", i), { n_embd }, flags | TENSOR_NOT_REQUIRED);
+        }
+    }
+}
+
+std::unique_ptr<llm_graph_context> llama_model_glm4::build_arch_graph(const llm_graph_params & params) const {
+    return std::make_unique<graph>(*this, params);
+}
+
+llama_model_glm4::graph::graph(const llama_model & model, const llm_graph_params & params) : llm_graph_context(params) {
    const int64_t n_embd_head = hparams.n_embd_head_v();

    GGML_ASSERT(n_embd_head == hparams.n_embd_head_k());