初始化项目，由ModelHub XC社区提供模型

Model: prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF Source: Original Platform
2026-06-22 07:07:17 +08:00
commit 3db218bfbc
9 changed files with 321 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,43 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz 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
+malaysian-mistral-7b-32k-instructions-f16.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q4_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q4_K_S.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q2_K.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q3_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q5_K_M.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.Q6_K.gguf filter=lfs diff=lfs merge=lfs -text
+malaysian-mistral-7b-32k-instructions.fp16.gguf filter=lfs diff=lfs merge=lfs -text
--- a/README.md
+++ b/README.md
@@ -0,0 +1,257 @@
+---
+base_model: mesolitica/malaysian-mistral-7b-32k-instructions
+inference: false
+model_creator: mesolitica
+model_name: Malaysian Mistral 7B 32k Instructions
+model_type: mistral
+pipeline_tag: text-generation
+prompt_template: |
+  <s>[INST] {prompt} [/INST]
+quantized_by: prsyahmi
+tags:
+- finetuned
+language:
+- ms
+---
+<!-- markdownlint-disable MD041 -->
+
+<!-- header start --><!-- header end -->
+
+# Malaysian Mistral 7B 32k Instructions - GGUF
+- Model creator: [mesolotica](https://huggingface.co/mesolitica)
+- Original model: [Malaysian Mistral 7B 32k Instructions](https://huggingface.co/mesolitica/malaysian-mistral-7b-32k-instructions)
+
+<!-- description start -->
+## Pengenalan
+Repo ini mengandungi model berformat GGUF untuk [mesolitica's Malaysian Mistral 7B 32k Instructions](https://huggingface.co/mesolitica/malaysian-mistral-7b-32k-instructions).
+
+GGUF adalah format kepada llama.cpp yang dibangunkan menggunakan C/C++ dimana pergantungan dengan software/library lain kurang menjadikan aplikasi ini ringan berbanding rata-rata aplikasi python.
+
+<!-- description end -->
+
+<!-- prompt-template start -->
+## Prompt template: Mistral
+```
+<s>[INST] {prompt} [/INST]
+
+```
+
+<!-- prompt-template end -->
+
+
+<!-- README_GGUF.md-provided-files start -->
+## Fail yang diberikan
+| Nama | Kaedah Quant | Saiz Fail |
+| ---- | ---- | ---- |
+| [malaysian-mistral-7b-32k-instructions.Q2_K.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q2_K.gguf) | Q2_K | 2.86 GB |
+| [malaysian-mistral-7b-32k-instructions.Q3_K_M.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q3_K_M.gguf) | Q3_K_M | 3.27 GB |
+| [malaysian-mistral-7b-32k-instructions.Q4_K_S.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q4_K_S.gguf) | Q4_K_S | 3.86 GB |
+| [malaysian-mistral-7b-32k-instructions.Q4_K_M.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q4_K_M.gguf) | Q4_K_M | 4.06 GB |
+| [malaysian-mistral-7b-32k-instructions.Q5_K_M.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q5_K_M.gguf) | Q5_K_M | 4.77 GB |
+| [malaysian-mistral-7b-32k-instructions.Q6_K.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.Q6_K.gguf) | Q6_K | 5.53 GB |
+| [malaysian-mistral-7b-32k-instructions.fp16.gguf](https://huggingface.co/prsyahmi/malaysian-mistral-7b-32k-instructions-GGUF/blob/main/malaysian-mistral-7b-32k-instructions.fp16.gguf) | FP16 | 13.5 GB |
+
+<!-- README_GGUF.md-provided-files end -->
+
+## Penghargaan
+Terima kasih kepada Husein Zolkepli dan keseluruhan team [mesolotica](https://huggingface.co/mesolitica)!
+
+Atas jasa mereka, kita dapat menggunakan atau mencuba AI peringkat tempatan.
+
+<!-- footer end -->
+
+-------
+
+<!-- original-model-card start -->
+# Full Parameter Finetuning 7B 32768 context length Mistral on Malaysian instructions dataset
+
+README at https://github.com/mesolitica/malaya/tree/5.1/session/mistral#instructions-7b-16384-context-length
+
+We use exact Mistral Instruct chat template.
+
+WandB, https://wandb.ai/mesolitica/fpf-mistral-7b-hf-instructions-16k?workspace=user-husein-mesolitica
+
+## how-to
+
+```python
+from transformers import AutoTokenizer, AutoModelForCausalLM, BitsAndBytesConfig
+import torch
+import json
+
+def parse_mistral_chat(messages, function_call = None):
+
+    user_query = messages[-1]['content']
+
+    users, assistants = [], []
+    for q in messages[:-1]:
+        if q['role'] == 'user':
+            users.append(q['content'])
+        elif q['role'] == 'assistant':
+            assistants.append(q['content'])
+
+    texts = ['<s>']
+    
+    if function_call:
+        fs = []
+        for f in function_call:
+            f = json.dumps(f, indent=4)
+            fs.append(f)
+        fs = '\n\n'.join(fs)
+        texts.append(f'\n[FUNCTIONCALL]\n{fs}\n')
+        
+    for u, a in zip(users, assistants):
+        texts.append(f'[INST] {u.strip()} [/INST] {a.strip()}</s>')
+
+    texts.append(f'[INST] {user_query.strip()} [/INST]')
+    prompt = ''.join(texts).strip()
+    return prompt
+
+TORCH_DTYPE = 'bfloat16'
+nf4_config = BitsAndBytesConfig(
+    load_in_4bit=True,
+    bnb_4bit_quant_type='nf4',
+    bnb_4bit_use_double_quant=True,
+    bnb_4bit_compute_dtype=getattr(torch, TORCH_DTYPE)
+)
+
+tokenizer = AutoTokenizer.from_pretrained('mesolitica/malaysian-mistral-7b-32k-instructions')
+model = AutoModelForCausalLM.from_pretrained(
+    'mesolitica/malaysian-mistral-7b-32k-instructions',
+    use_flash_attention_2 = True,
+    quantization_config = nf4_config
+)
+
+messages = [
+    {'role': 'user', 'content': 'kwsp tu apa'}
+]
+prompt = parse_mistral_chat(messages)
+inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda')
+generate_kwargs = dict(
+    inputs,
+    max_new_tokens=1024,
+    top_p=0.95,
+    top_k=50,
+    temperature=0.9,
+    do_sample=True,
+    num_beams=1,
+)
+r = model.generate(**generate_kwargs)
+tokenizer.decode(r[0])
+```
+
+```text
+<s> [INST] kwsp tu apa [/INST]KWSP bermaksud Kumpulan Wang Simpanan Pekerja. Ia adalah sebuah institusi simpanan persaraan yang ditubuhkan oleh Kementerian Kewangan Malaysia untuk tujuan mengumpul simpanan ahli untuk dibayar pada umur persaraan, penuh atau penuh persaraan penuh. KWSP ditubuhkan pada tahun 1951 dan mula beroperasi pada tahun 1952. KWSP adalah salah satu institusi simpanan persaraan terbesar di dunia, dengan pangkalan ahli sekitar 14 juta ahli.</s>
+```
+
+```python
+messages = [
+    {'role': 'user', 'content': 'camne nak kill pid dalam linux'}
+]
+prompt = parse_mistral_chat(messages)
+inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda')
+generate_kwargs = dict(
+    inputs,
+    max_new_tokens=1024,
+    top_p=0.95,
+    top_k=50,
+    temperature=0.9,
+    do_sample=True,
+    num_beams=1,
+)
+r = model.generate(**generate_kwargs)
+print(tokenizer.decode(r[0]))
+```
+
+```text
+<s> [INST] camne nak kill pid dalam linux [/INST]Untuk membunuh pid dalam Linux, ikuti langkah-langkah berikut:
+
+1. Cari pid menggunakan perintah `ps aux | grep 'pid'`. Pilih pid yang anda ingin bunuh.
+
+2. Gunakan perintah `kill` diikuti dengan nombor pid. Sebagai contoh, jika anda ingin membunuh pid dengan nombor 1234, gunakan perintah `kill -1234`.
+
+3. Sahkan pembunuhan dengan menjalankan perintah `kill -id`. Jika perintah tidak mengembalikan sebarang ralat, maka pid telah dibunuh dengan berjaya.
+
+4. Anda juga boleh menggunakan perintah `kill -s` untuk membunuh semua pid dengan nama atau atribut tertentu.</s>
+```
+
+```python
+f = {
+    'name': 'parse_entities',
+    'description': 'extract entities from the text',
+    'parameters': {
+        'type': 'object',
+        'properties': {
+            'drink': {
+                'type': 'string',
+                'description': 'drink name',
+            },
+            'event': {
+                'type': 'string',
+                'description': 'event name',
+            },
+            'person_name': {
+                'type': 'string',
+                'description': 'person name',
+            }
+        },
+        'required': [
+            'drink',
+            'event',
+            'person_name'
+        ]
+    }
+}
+messages = [
+    {'role': 'user', 'content': 'nama saya husein bin zolkepli, saya sekarang berada di jomheboh 2023 sambil minum starbucks'}
+]
+prompt = parse_mistral_chat(messages, function_call = [f])
+inputs = tokenizer([prompt], return_tensors='pt', add_special_tokens=False).to('cuda')
+generate_kwargs = dict(
+    inputs,
+    max_new_tokens=128,
+    top_p=0.95,
+    top_k=50,
+    temperature=0.9,
+    do_sample=True,
+    num_beams=1,
+)
+r = model.generate(**generate_kwargs)
+print(tokenizer.decode(r[0]))
+```
+
+```text
+<s> 
+[FUNCTIONCALL]
+{
+    "name": "parse_entities",
+    "description": "extract entities from the text",
+    "parameters": {
+        "type": "object",
+        "properties": {
+            "drink": {
+                "type": "string",
+                "description": "drink name"
+            },
+            "event": {
+                "type": "string",
+                "description": "event name"
+            },
+            "person_name": {
+                "type": "string",
+                "description": "person name"
+            }
+        },
+        "required": [
+            "drink",
+            "event",
+            "person_name"
+        ]
+    }
+}
+[INST] nama saya husein bin zolkepli, saya sekarang berada di jomheboh 2023 sambil minum starbucks [/INST] <functioncall> {"name": "parse_entities", "arguments": '{
+  "drink": "Starbucks",
+  "event": "Jom Heboh 2023",
+  "person_name": "Husein Bin Zolkepli"
+}'}</s>
+```
+<!-- original-model-card end -->
--- a/malaysian-mistral-7b-32k-instructions.Q2_K.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q2_K.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bef7ddae7e34d758457ef480aec26b0fb780038db98fd7f1d9d56f0104fea4d0
+size 3083097728
--- a/malaysian-mistral-7b-32k-instructions.Q3_K_M.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q3_K_M.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:af51b6d5c2cd08ca5a0bbb87fcc9cdc43791dbfb6ed5ce73359bcebc7d6e0210
+size 3518985856
--- a/malaysian-mistral-7b-32k-instructions.Q4_K_M.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q4_K_M.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:15342fa62764c9890d8ee4702dc55ac00490d31e94f03a94285911cffd24abbc
+size 4368438912
--- a/malaysian-mistral-7b-32k-instructions.Q4_K_S.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q4_K_S.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1c3465f18d2ab6c4ec2957bae00a31100d0046901f4bd13641654c46d068f4bd
+size 4140373632
--- a/malaysian-mistral-7b-32k-instructions.Q5_K_M.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q5_K_M.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:18713c61ca8b8082bb515fb1aa51b36550fc17ec6b1f741009ab4f395fa84c69
+size 5131409024
--- a/malaysian-mistral-7b-32k-instructions.Q6_K.gguf
+++ b/malaysian-mistral-7b-32k-instructions.Q6_K.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f8b19a5cb08aa00aecb859b80490212b459afa035ddc50587748135d0797fd87
+size 5942064768
--- a/malaysian-mistral-7b-32k-instructions.fp16.gguf
+++ b/malaysian-mistral-7b-32k-instructions.fp16.gguf
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1108ac7504fd9b6edfa0d7f4b61ef85f6c1a533ece046b059152ed2f1d30305e
+size 14484731488