Upload 51 files

.gitattributes

@@ -33,3 +33,43 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+main_axcl_aarch64 filter=lfs diff=lfs merge=lfs -text
+main_axcl_x86 filter=lfs diff=lfs merge=lfs -text
+main_prefill filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l0_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l1_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l10_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l11_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l12_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l13_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l14_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l15_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l16_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l17_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l18_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l19_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l2_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l20_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l21_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l22_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l23_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l24_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l25_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l26_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l27_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l28_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l29_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l3_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l30_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l31_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l32_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l33_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l34_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l35_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l4_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l5_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l6_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l7_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l8_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l9_together.axmodel filter=lfs diff=lfs merge=lfs -text
+qwen2.5-3b-gptq-int4-ax650/qwen2_post.axmodel filter=lfs diff=lfs merge=lfs -text

main_axcl_aarch64

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb111fc00c54abb6142a8f44df087bf104c8150a1cefa6be55c6b174b932c4ec
+size 999008

main_axcl_x86

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cba0be8df523e351789fcfa745772423096f52a3b0a760f8d8c9f5b8bb2ec82
+size 1022384

main_prefill

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:90026f1e071aef781ccc9bc313db3805749b3e234bf03d578c9f97ae2e60abd2
+size 953544

post_config.json

@@ -0,0 +1,14 @@
+{
+    "enable_temperature" : true,
+    "temperature" : 0.9,
+    
+    "enable_repetition_penalty" : false,
+    "repetition_penalty" : 1.2,
+    "penalty_window" : 20,
+
+    "enable_top_p_sampling" : false,
+    "top_p" : 0.8,
+
+    "enable_top_k_sampling" : true,
+    "top_k" : 10
+}

qwen2.5-3b-gptq-int4-ax650/model.embed_tokens.weight.bfloat16.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03bd8df1218b5bcb8439e6a0d10d49e2cbc55e002f6552401e3e0cc71291fe87
+size 622329856

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l0_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ee40a01df23502c990c4b5591bfe2416bd32b4298b395c16c835656e4cdb32ac
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l10_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89d5682b484edb15620990b8a624aafc65f67c46230392bc8d502097e565bd2b
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l11_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cafef780a1ace888e393584087e433802415f732163a1164949aa338d59e819b
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l12_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7bfc3ac12b466ccddf795dea04954bbebaba7e335da7dd59ba609dcce809bc6b
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l13_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a48f67d0c05f27d65cc362f7f52b47cf30d24c5401c86b4eb1c5e7205049595
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l14_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf11b09aaa793fff665ceebd5d4e0012a49073c2bac014aeff562f001cd2a48f
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l15_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bafe532f7c42229b960fad6c8154bab9662e31f2afc755069e3227d6243dfa34
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l16_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc07aed4e27128f723fa8eb8cee78a687044e29b7ad3ea6205a568983b23bb79
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l17_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:19898373af7c40589a87979496fbb1fbdf4d9807137f10ffcef5e6345734b328
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l18_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f0b9fe93c3b894f2858c4b718b7d2a6783f1effb730c400a681bd863bb947b35
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l19_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b329ba7920daee20a0302ea1844adf0d54125c10515d7d97ba85370ea45b90d0
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l1_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb7d8a79dfc746ec070e66fbfac38ff88443a924ce4292977801db6a6dd4794
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l20_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6f584853938bcefc233bc232523018e6b2130e1d2232a051c89cab92ef4b0f0
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l21_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:16be46816d2392feb977b42f010b50463ad4e1625a413c6ed416d91fdb2de78e
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l22_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:afb1b1880597ab191866099fd1894b0d063341e07cd2acc7546bc5e71c74cefb
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l23_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d6be83906546892ef040e4087e532ad0cafae558d2902b7a93fcaa9d89a6c426
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l24_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ab438a8f0ddcbdcd26f4741e381d2b5a3c6d3a647a201974e5ca49fd77c4bae
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l25_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8944516777a78114a4adf67eb7b527aeb0c490d8d5a09d9abd7c4d608621b01
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l26_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:194a171e68496f6360faf0342edb04febe7f3deb6c5d22717d7b7defdb50f0df
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l27_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:662b81eb7719adefa8d4ba7b37b08004d97b855a830f20628aad283fdcf2d66f
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l28_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b6de1a5e7543b55038dbfa1f4ebffe6f013286dc5af998f04f739f23562ff9a
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l29_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:de15ca4d7f5129d7227fb3d402685ca79455134a9d0d771a109276d1cd2f985d
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l2_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6927920bc2ceecd57a56f401919015f549cacd763874bc4d7dd4da982d19bac6
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l30_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc0d38f30d83ed2f298ff9e461d109905539268274ecd7464640b9d50879ba42
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l31_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2facde6f440be3eafcbd8d29523f721d7d56f12dc324903aa7389cd2d6e61d7c
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l32_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa3f94be152842621fa4b9d05e748ff710219c5bdb881a2b6ab86ece5d064238
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l33_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d517373ff377fe01ee9dd28ee717be166ab06e0d3004f067130533bf84844ae1
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l34_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c43e56441ae29f207b2ee63a0b4707e5d73d501cd5b07acc3bb3cfc9a4df6502
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l35_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8f8088c192bb4d842a03826d1fe2fc01bd2d2ec694b0d73394cc90cf99737b32
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l3_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:426c8e7b1def9bff1290c63ccbbeaf2e75e59400efa81149ca122da401dfc4a6
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l4_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:072a68341f146e0d9351e30f79e52eedf6ba1e66d8915a32cff6a2ea5bfe4a26
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l5_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e4f6ec671e2f67537e410852f46223fea9f4f3b453405de2ac8548bcc60a838
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l6_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32752e328d1e29e6e1560d9ef95a8f3e0fa7ccfab9c538b682ceba304d94979d
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l7_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e83941de69c199e8e10f8e145fa841a8fcec72f7e9f50312995f65a8a216421
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l8_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d8e311e8efe3023bcbff557eefd9afa0824c0b502b5bddd482e0808bfba05b5
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l9_together.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:047d8d873908dd03744513afc732d3b39b6d8b38736b453e84f564d4a3d41e38
+size 45856164

qwen2.5-3b-gptq-int4-ax650/qwen2_post.axmodel

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:803ad252d1bc5712c50464a9da9c4ff2a47f6d37a1f3702c3716b0bcb9aee7a6
+size 339625308

qwen2.5_tokenizer.py

@@ -0,0 +1,133 @@
+from transformers import AutoTokenizer, PreTrainedTokenizerFast
+from http.server import HTTPServer, BaseHTTPRequestHandler
+import json
+import argparse
+
+
+class Tokenizer_Http():
+
+    def __init__(self):
+        model_id = "qwen2.5_tokenizer"
+        self.tokenizer = AutoTokenizer.from_pretrained(model_id)
+
+    def encode(self, prompt):
+        messages = [
+            {"role": "system", "content": "You are Qwen, created by Alibaba Cloud. You are a helpful assistant."},
+            {"role": "user", "content": prompt}
+        ]
+        text = self.tokenizer.apply_chat_template(
+            messages,
+            tokenize=False,
+            add_generation_prompt=True
+        )
+        print(text)
+        token_ids = self.tokenizer.encode(text)
+        return token_ids
+
+    def decode(self, token_ids):
+        return self.tokenizer.decode(token_ids)
+
+    @property
+    def bos_id(self):
+        return self.tokenizer.bos_token_id
+
+    @property
+    def eos_id(self):
+        return self.tokenizer.eos_token_id
+    
+    @property
+    def bos_token(self):
+        return self.tokenizer.bos_token
+
+    @property
+    def eos_token(self):
+        return self.tokenizer.eos_token
+
+
+tokenizer = Tokenizer_Http()
+
+print(tokenizer.bos_id, tokenizer.bos_token, tokenizer.eos_id, tokenizer.eos_token)
+print(tokenizer.encode("hello world"))
+
+
+class Request(BaseHTTPRequestHandler):
+    #通过类继承，新定义类
+    timeout = 5
+    server_version = 'Apache'
+
+    def do_GET(self):
+        print(self.path)
+        #在新类中定义get的内容（当客户端向该服务端使用get请求时，本服务端将如下运行）
+        self.send_response(200)
+        self.send_header("type", "get")  #设置响应头，可省略或设置多个
+        self.end_headers()
+
+        if self.path == '/bos_id':
+            bos_id = tokenizer.bos_id
+            # print(bos_id)
+            # to json
+            if bos_id is None:
+                msg = json.dumps({'bos_id': -1})
+            else:
+                msg = json.dumps({'bos_id': bos_id})
+        elif self.path == '/eos_id':
+            eos_id = tokenizer.eos_id
+            if eos_id is None:
+                msg = json.dumps({'eos_id': -1})
+            else:
+                msg = json.dumps({'eos_id': eos_id})
+        else:
+            msg = 'error'
+
+        print(msg)
+        msg = str(msg).encode()  #转为str再转为byte格式
+
+        self.wfile.write(msg)  #将byte格式的信息返回给客户端
+
+    def do_POST(self):
+        #在新类中定义post的内容（当客户端向该服务端使用post请求时，本服务端将如下运行）
+        data = self.rfile.read(int(
+            self.headers['content-length']))  #获取从客户端传入的参数（byte格式）
+        data = data.decode()  #将byte格式转为str格式
+
+        self.send_response(200)
+        self.send_header("type", "post")  #设置响应头，可省略或设置多个
+        self.end_headers()
+
+        if self.path == '/encode':
+            req = json.loads(data)
+            prompt = req['text']
+
+            token_ids = tokenizer.encode(prompt)
+            if token_ids is None:
+                msg = json.dumps({'token_ids': -1})
+            else:
+                msg = json.dumps({'token_ids': token_ids})
+
+        elif self.path == '/decode':
+            req = json.loads(data)
+            token_ids = req['token_ids']
+            text = tokenizer.decode(token_ids)
+            if text is None:
+                msg = json.dumps({'text': ""})
+            else:
+                msg = json.dumps({'text': text})
+        else:
+            msg = 'error'
+        print(msg)
+        msg = str(msg).encode()  #转为str再转为byte格式
+
+        self.wfile.write(msg)  #将byte格式的信息返回给客户端
+
+
+if __name__ == "__main__":
+
+    args = argparse.ArgumentParser()
+    args.add_argument('--host', type=str, default='localhost')
+    args.add_argument('--port', type=int, default=8080)
+    args = args.parse_args()
+
+    host = (args.host, args.port)  #设定地址与端口号，'localhost'等价于'127.0.0.1'
+    print('http://%s:%s' % host)
+    server = HTTPServer(host, Request)  #根据地址端口号和新定义的类，创建服务器实例
+    server.serve_forever()  #开启服务

qwen2.5_tokenizer/tokenizer_config.json

@@ -0,0 +1,207 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- messages[0]['content'] }}\n    {%- else %}\n        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}\n    {%- endif %}\n    {{- \"\\n\\n# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0]['role'] == 'system' %}\n        {{- '<|im_start|>system\\n' + messages[0]['content'] + '<|im_end|>\\n' }}\n    {%- else %}\n        {{- '<|im_start|>system\\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- for message in messages %}\n    {%- if (message.role == \"user\") or (message.role == \"system\" and not loop.first) or (message.role == \"assistant\" and not message.tool_calls) %}\n        {{- '<|im_start|>' + message.role + '\\n' + message.content + '<|im_end|>' + '\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role }}\n        {%- if message.content %}\n            {{- '\\n' + message.content }}\n        {%- endif %}\n        {%- for tool_call in message.tool_calls %}\n            {%- if tool_call.function is defined %}\n                {%- set tool_call = tool_call.function %}\n            {%- endif %}\n            {{- '\\n<tool_call>\\n{\"name\": \"' }}\n            {{- tool_call.name }}\n            {{- '\", \"arguments\": ' }}\n            {{- tool_call.arguments | tojson }}\n            {{- '}\\n</tool_call>' }}\n        {%- endfor %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {{- message.content }}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null,
+  "add_bos_token": false
+}

run_qwen2.5_3b_gptq_int4_ax650.sh

@@ -0,0 +1,14 @@
+./main_prefill \
+--template_filename_axmodel "qwen2.5-3b-gptq-int4-ax650/qwen2_p128_l%d_together.axmodel" \
+--axmodel_num 36 \
+--tokenizer_type 2 \
+--filename_tokenizer_model "http://127.0.0.1:12345" \
+--bos 0 --eos 0 \
+--filename_post_axmodel "qwen2.5-3b-gptq-int4-ax650/qwen2_post.axmodel" \
+--filename_tokens_embed "qwen2.5-3b-gptq-int4-ax650/model.embed_tokens.weight.bfloat16.bin" \
+--tokens_embed_num 151936 \
+--tokens_embed_size 2048 \
+--use_mmap_load_embed 1 \
+--live_print 1 \
+--continue 1 \
+--prompt "$1"