add glm edge chat model

convert_hf_to_gguf.py

@@ -625,7 +625,7 @@ class Model:
         if chkhsh == "7967bfa498ade6b757b064f31e964dddbb80f8f9a4d68d4ba7998fcf281c531a":
             # ref: https://huggingface.co/jinaai/jina-embeddings-v2-base-code
             res = "jina-v2-code"
-        if chkhsh == "b6e8e1518dc4305be2fe39c313ed643381c4da5db34a98f6a04c093f8afbe99b":
+        if chkhsh == "b6e8e1518dc4305be2fe39c313ed643381c4da5db34a98f6a04c093f8afbe99b" or chkhsh == "81d72c7348a9f0ebe86f23298d37debe0a5e71149e29bd283904c02262b27516":
             # ref: https://huggingface.co/THUDM/glm-4-9b-chat
             res = "chatglm-bpe"
         if chkhsh == "7fc505bd3104ca1083b150b17d088b59534ede9bde81f0dd2090967d7fe52cee":
@@ -3989,7 +3989,14 @@ class ChatGLMModel(Model):
         self.gguf_writer.add_head_count_kv(n_head_kv)
         self.gguf_writer.add_layer_norm_rms_eps(self.hparams["layernorm_epsilon"])
         self.gguf_writer.add_file_type(self.ftype)
-        self.gguf_writer.add_rope_dimension_count(64)
+        if "attention_dim" in self.hparams:
+            rope_dim = self.hparams["attention_dim"]
+        else:
+            rope_dim = self.hparams["hidden_size"] // self.hparams["num_attention_heads"]
+        if "THUDM/glm4-nano" in self.hparams.get("_name_or_path", "") or "THUDM/glm4-mini" in self.hparams.get("_name_or_path", ""):
+            self.gguf_writer.add_rope_dimension_count(rope_dim)
+        else:
+            self.gguf_writer.add_rope_dimension_count(int(rope_dim/2))
         self.gguf_writer.add_add_bos_token(False)
         rope_freq = 10000
         if "rope_ratio" in self.hparams:

src/llama.cpp

@@ -5912,8 +5912,20 @@ static void llm_load_hparams(
             {
                 ml.get_key(LLM_KV_ATTENTION_LAYERNORM_RMS_EPS, hparams.f_norm_rms_eps);
                 switch (hparams.n_layer) {
-                    case 28: model.type = e_model::MODEL_6B; break;
+                    case 28: {
-                    case 40: model.type = e_model::MODEL_9B; break;
+                        if(hparams.n_head(0)==16){
+                            model.type = e_model::MODEL_1_6B;
+                        }else{
+                            model.type = e_model::MODEL_6B;
+                        }
+                    }break;
+                    case 40:{
+                        if(hparams.n_head(0)==24){
+                            model.type = e_model::MODEL_4B;
+                        }else{
+                            model.type = e_model::MODEL_9B;
+                        }
+                    } break;
                     default: model.type = e_model::MODEL_UNKNOWN;
                 }
             } break;
@@ -8859,7 +8871,7 @@ static bool llm_load_tensors(
                         layer.attn_norm = create_tensor(tn(LLM_TENSOR_ATTN_NORM, "weight", i), {n_embd}, 0);
 
                         layer.wqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "weight", i), {n_embd, n_embd + 2*n_embd_gqa}, 0);
-                        layer.bqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "bias", i),   {n_embd + 2*n_embd_gqa}, 0);
+                        layer.bqkv = create_tensor(tn(LLM_TENSOR_ATTN_QKV, "bias", i),   {n_embd + 2*n_embd_gqa}, llama_model_loader::TENSOR_NOT_REQUIRED);
 
                         layer.wo   = create_tensor(tn(LLM_TENSOR_ATTN_OUT, "weight", i), {n_embd, n_embd}, 0);
 
@@ -15722,8 +15734,10 @@ struct llm_build_context {
                 cur = llm_build_lora_mm(lctx, ctx0, model.layers[il].wqkv, cur);
                 cb(cur, "wqkv", il);
 
+                if(model.layers[il].bqkv){
                     cur = ggml_add(ctx0, cur, model.layers[il].bqkv);
                     cb(cur, "bqkv", il);
+                }
 
                 Qcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd,     n_tokens, cur->nb[1], 0*sizeof(float)*(n_embd)));
                 Kcur = ggml_cont(ctx0, ggml_view_2d(ctx0, cur, n_embd_gqa, n_tokens, cur->nb[1], 1*sizeof(float)*(n_embd)));
@@ -21742,6 +21756,14 @@ static int32_t llama_chat_apply_template_internal(
         if (add_ass) {
             ss << "<|assistant|>";
         }
+    } else if(tmpl == "glm-edge" || tmpl_contains("<|assistant|>")){
+        for (auto message : chat) {
+            std::string role(message->role);
+            ss << "<|" << role << "|>" << "\n" << message->content;
+        }
+        if (add_ass) {
+            ss << "<|assistant|>";
+        }
     } else if (tmpl == "minicpm" || tmpl_contains(LU8("<用户>"))) {
         // MiniCPM-3B-OpenHermes-2.5-v2-GGUF
         for (auto message : chat) {

tests/test-chat-template.cpp

@@ -61,6 +61,8 @@ int main(void) {
         "{% for message in messages %}{% if loop.first %}[gMASK]sop<|{{ message['role'] }}|>\n {{ message['content'] }}{% else %}<|{{ message['role'] }}|>\n {{ message['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}",
         // ChatGLM4
         u8"[gMASK]<sop>{% for item in messages %}{% if item['tools'] is defined %}<|system|>\n你是一个名为 ChatGLM 的人工智能助手。你是基于智谱AI训练的语言模型 GLM-4 模型开发的，你的任务是针对用户的问题和要求提供适当的答复和支持。\n\n# 可用工具{% set tools = item['tools'] %}{% for tool in tools %}{% if tool['type'] == 'function' %}\n\n## {{ tool['function']['name'] }}\n\n{{ tool['function'] | tojson(indent=4) }}\n......{% endif %}{% endfor %}{% endif %}{% if item['content'] %}<|{{ item['role'] }}|>{{ item['metadata'] }}\n{{ item['content'] }}{% endif %}{% endfor %}{% if add_generation_prompt %}<|assistant|>{% endif %}",
+        // GLM-edge
+        "{% for message in messages %}{{'<|' + message['role'] + '|>' + '\n' + message['content'] +'\n' }}{% endfor %}{% if add_generation_prompt and messages[-1]['role'] != 'assistant' %}{{- '<|assistant|>\n' -}}{% endif %}",
         // MiniCPM-3B-OpenHermes-2.5-v2-GGUF
         u8"{% for message in messages %}{% if message['role'] == 'user' %}{{'<用户>' + message['content'].strip() + '<AI>'}}{% else %}{{message['content'].strip()}}{% endif %}{% endfor %}",
         // DeepSeek-V2
@@ -107,6 +109,8 @@ int main(void) {
         "[gMASK]sop<|system|>\n You are a helpful assistant<|user|>\n Hello<|assistant|>\n Hi there<|user|>\n Who are you<|assistant|>\n    I am an assistant   <|user|>\n Another question<|assistant|>",
         // ChatGLM4
         "[gMASK]<sop><|system|>\nYou are a helpful assistant<|user|>\nHello<|assistant|>\nHi there<|user|>\nWho are you<|assistant|>\n   I am an assistant   <|user|>\nAnother question<|assistant|>",
+        // GLM-Edge
+        "<|system|>\nYou are a helpful assistant<|user|>\nHello<|assistant|>\nHi there<|user|>\nWho are you<|assistant|>\n   I am an assistant   <|user|>\nAnother question<|assistant|>",
         // MiniCPM-3B-OpenHermes-2.5-v2-GGUF
         u8"You are a helpful assistant<用户>Hello<AI>Hi there<用户>Who are you<AI>I am an assistant<用户>Another question<AI>",
         // DeepSeek-V2