Update src/streamlit_app.py

src/streamlit_app.py

@@ -203,86 +203,315 @@ LLM_DATABASE = {
 
         ],
         "code": [
-            {"name": "CodeGen-2B", "size": "1.8GB", "description": "Salesforce code model", "parameters": "2B", "context": "2K"},
-            {"name": "StarCoder-1B", "size": "1.1GB", "description": "BigCode project", "parameters": "1B", "context": "8K"}
-        ],
+            
+  { "name": "CodeGen-2B", "size": "1.8GB", "description": "Salesforce's code generation model" },
+  { "name": "StarCoder-1B", "size": "1.1GB", "description": "BigCode's programming assistant" },
+  { "name": "InCoder-1B", "size": "1.0GB", "description": "Facebook's code infilling model" },
+  { "name": "PolyCoder-2.7B", "size": "2.7GB", "description": "Carnegie Mellon's code model" },
+  { "name": "CodeParrot-small", "size": "1.5GB", "description": "HuggingFace's Python code model" },
+  { "name": "SantaCoder-1.1B", "size": "1.1GB", "description": "BigCode's multilingual code model" },
+  { "name": "GPT-Code-2B", "size": "2.0GB", "description": "Code-specialized GPT variant" },
+  { "name": "AlphaCode-2B", "size": "2.2GB", "description": "DeepMind's programming model" },
+  { "name": "Codex-2B", "size": "2.0GB", "description": "OpenAI's code generation model" },
+  { "name": "TabNine-2B", "size": "2.1GB", "description": "AI code completion assistant" }
+
+     ],
         "chat": [
-            {"name": "Alpaca-3B", "size": "2.0GB", "description": "Stanford's instruction model", "parameters": "3B", "context": "2K"},
-            {"name": "Vicuna-3B", "size": "2.1GB", "description": "ChatGPT-style training", "parameters": "3B", "context": "2K"}
+           
+  { "name": "Alpaca-3B", "size": "2.0GB", "description": "Stanford's instruction-following model" },
+  { "name": "Vicuna-3B", "size": "2.1GB", "description": "UC Berkeley's chat model" },
+  { "name": "Dolly-3B", "size": "2.2GB", "description": "Databricks' instruction-tuned model" },
+  { "name": "OpenAssistant-3B", "size": "2.3GB", "description": "LAION's assistant model" },
+  { "name": "StableVicuna-3B", "size": "2.1GB", "description": "Stable version of Vicuna" },
+  { "name": "MPT-3B-Chat", "size": "2.0GB", "description": "MosaicML's chat variant" },
+  { "name": "RedPajama-Chat-3B", "size": "2.1GB", "description": "Together AI's chat model" },
+  { "name": "OpenChatKit-3B", "size": "2.2GB", "description": "Together AI's open chat model" },
+  { "name": "Koala-3B", "size": "2.0GB", "description": "UC Berkeley's dialogue model" },
+  { "name": "Guanaco-3B", "size": "2.1GB", "description": "QLoRA fine-tuned model" }
+
+
         ]
     },
+    "reasoning": [
+  { "name": "WizardMath-7B", "size": "4.0GB", "description": "Mathematical reasoning specialist" },
+  { "name": "MAmmoTH-7B", "size": "4.1GB", "description": "Mathematical reasoning model" },
+  { "name": "MetaMath-7B", "size": "3.9GB", "description": "Mathematical problem solver" },
+  { "name": "Abel-7B", "size": "4.0GB", "description": "Advanced reasoning capabilities" },
+  { "name": "Orca-2-7B", "size": "4.1GB", "description": "Microsoft's reasoning specialist" }
+]
+},
     "moderate_low": {  # 5-6GB
         "general": [
-            {"name": "Phi-2", "size": "5.2GB", "description": "Microsoft's 2.7B model", "parameters": "2.7B", "context": "2K"},
-            {"name": "Gemma-7B-it", "size": "4.2GB", "description": "Google instruction tuned", "parameters": "7B", "context": "8K"},
-            {"name": "Mistral-7B-v0.1", "size": "4.1GB", "description": "Mistral AI base model", "parameters": "7B", "context": "8K"},
-            {"name": "Llama-2-7B", "size": "4.0GB", "description": "Meta's foundation model", "parameters": "7B", "context": "4K"}
-        ],
+            
+  { "name": "Phi-2", "size": "5.2GB", "description": "Microsoft's advanced 2.7B parameter model" },
+  { "name": "Gemma-7B", "size": "4.2GB", "description": "Google's efficient 7B model" },
+  { "name": "Mistral-7B-v0.1", "size": "4.1GB", "description": "Mistral AI's foundation model" },
+  { "name": "OpenLLaMA-7B", "size": "4.0GB", "description": "Open source 7B language model" },
+  { "name": "MPT-7B", "size": "4.3GB", "description": "MosaicML's transformer model" },
+  { "name": "Falcon-7B", "size": "4.1GB", "description": "TII's instruction model" },
+  { "name": "Pythia-6.9B", "size": "6.9GB", "description": "EleutherAI's large training model" },
+  { "name": "BLOOM-7B", "size": "7.0GB", "description": "BigScience's multilingual foundation model" },
+  { "name": "OLMo-7B", "size": "4.2GB", "description": "Allen AI's open language model" },
+  { "name": "Llama-7B", "size": "4.0GB", "description": "Meta's foundation model" },
+  { "name": "StableLM-7B", "size": "4.1GB", "description": "Stability AI's larger model" },
+  { "name": "RedPajama-7B", "size": "4.0GB", "description": "Together AI's 7B model" },
+  { "name": "OpenLLaMA-7B-v2", "size": "4.1GB", "description": "Improved OpenLLaMA version" },
+  { "name": "Vicuna-7B", "size": "3.9GB", "description": "UC Berkeley's 7B chat model" },
+  { "name": "Alpaca-7B", "size": "3.8GB", "description": "Stanford's instruction model" },
+  { "name": "GPT-NeoX-6B", "size": "6.0GB", "description": "EleutherAI's improved model" },
+  { "name": "OPT-6.7B", "size": "6.7GB", "description": "Meta's 6.7B parameter model" },
+  { "name": "T5-large", "size": "3.0GB", "description": "Large Text-to-Text Transfer" },
+  { "name": "FLAN-T5-large", "size": "3.2GB", "description": "Instruction-tuned T5 large" },
+  { "name": "UL2-base", "size": "4.0GB", "description": "Unified Language Learner base" }
+]
+
         "code": [
-            {"name": "CodeLlama-7B", "size": "3.8GB", "description": "Meta's code specialist", "parameters": "7B", "context": "16K"},
-            {"name": "StarCoder-7B", "size": "4.0GB", "description": "Code generation expert", "parameters": "7B", "context": "8K"}
+           
+  { "name": "CodeLlama-7B", "size": "3.8GB", "description": "Meta's specialized code model" },
+  { "name": "StarCoder-7B", "size": "4.0GB", "description": "Advanced code generation model" },
+  { "name": "SantaCoder-1.1B", "size": "1.2GB", "description": "Multilingual code model" },
+  { "name": "CodeGen-6B", "size": "6.0GB", "description": "Salesforce's larger code model" },
+  { "name": "CodeT5p-6B", "size": "6.2GB", "description": "Salesforce's code understanding model" },
+  { "name": "InCoder-6B", "size": "6.0GB", "description": "Facebook's large infilling model" },
+  { "name": "PolyCoder-6B", "size": "6.1GB", "description": "Carnegie Mellon's large code model" },
+  { "name": "AlphaCode-7B", "size": "4.0GB", "description": "DeepMind's competitive programming" },
+  { "name": "Codex-7B", "size": "4.1GB", "description": "OpenAI's advanced code model" },
+  { "name": "WizardCoder-7B", "size": "4.0GB", "description": "Microsoft's coding wizard" }
+
+
         ],
         "chat": [
-            {"name": "Zephyr-7B-beta", "size": "4.2GB", "description": "HuggingFace chat model", "parameters": "7B", "context": "32K"},
-            {"name": "Neural-Chat-7B", "size": "4.1GB", "description": "Intel optimized", "parameters": "7B", "context": "32K"}
+       
+  { "name": "Zephyr-7B-beta", "size": "4.2GB", "description": "HuggingFace's chat specialist" },
+  { "name": "Neural-Chat-7B", "size": "4.1GB", "description": "Intel's optimized chat model" },
+  { "name": "OpenChat-7B", "size": "4.0GB", "description": "High-quality conversation model" },
+  { "name": "Nous-Hermes-7B", "size": "4.1GB", "description": "NousResearch's assistant model" },
+  { "name": "StableBeluga-7B", "size": "4.2GB", "description": "Stability AI's chat model" },
+  { "name": "Llama-2-7B-Chat", "size": "3.9GB", "description": "Meta's chat-optimized model" },
+  { "name": "Vicuna-7B-v1.3", "size": "3.9GB", "description": "Improved Vicuna chat model" },
+  { "name": "WizardLM-7B", "size": "4.0GB", "description": "Microsoft's instruction model" },
+  { "name": "Orca-Mini-7B", "size": "4.1GB", "description": "Microsoft's reasoning model" },
+  { "name": "Samantha-7B", "size": "4.0GB", "description": "Eric Hartford's assistant model" }
+
+
         ]
     },
     "moderate": {  # 7-8GB
         "general": [
-            {"name": "Llama-2-7B-Chat", "size": "3.5GB", "description": "Meta's popular chat model", "parameters": "7B", "context": "4K"},
-            {"name": "Mistral-7B-Instruct-v0.2", "size": "4.1GB", "description": "Latest Mistral instruct", "parameters": "7B", "context": "32K"},
-            {"name": "Qwen-7B-Chat", "size": "4.0GB", "description": "Alibaba's multilingual", "parameters": "7B", "context": "32K"},
-            {"name": "Solar-10.7B-Instruct", "size": "5.8GB", "description": "Upstage's efficient model", "parameters": "10.7B", "context": "4K"}
-        ],
+            
+  { "name": "Llama-2-7B-Chat", "size": "3.5GB", "description": "Meta's popular chat model (4-bit)" },
+  { "name": "Mistral-7B-Instruct-v0.2", "size": "4.1GB", "description": "Latest Mistral instruction model" },
+  { "name": "Qwen-7B-Chat", "size": "4.0GB", "description": "Alibaba's multilingual model" },
+  { "name": "Baichuan2-7B-Chat", "size": "4.1GB", "description": "Chinese LLM with strong capabilities" },
+  { "name": "Yi-6B-Chat", "size": "3.8GB", "description": "01.AI's bilingual chat model" },
+  { "name": "InternLM-7B-Chat", "size": "4.0GB", "description": "Shanghai AI Lab's model" },
+  { "name": "ChatGLM3-6B", "size": "3.7GB", "description": "Tsinghua's latest chat model" },
+  { "name": "Aquila-7B", "size": "4.1GB", "description": "BAAI's Chinese-English model" },
+  { "name": "Skywork-13B", "size": "7.2GB", "description": "Kunlun's bilingual model" },
+  { "name": "Llama-2-7B", "size": "3.8GB", "description": "Meta's base foundation model" },
+  { "name": "Mistral-7B-v0.1", "size": "4.0GB", "description": "Original Mistral foundation" },
+  { "name": "Solar-10.7B", "size": "5.4GB", "description": "Upstage's efficient model" },
+  { "name": "Nous-Hermes-2-7B", "size": "4.0GB", "description": "NousResearch's improved model" },
+  { "name": "OpenHermes-2.5-7B", "size": "4.1GB", "description": "Teknium's assistant model" },
+  { "name": "Starling-LM-7B", "size": "4.0GB", "description": "Berkeley's RLAIF model" },
+  { "name": "Openchat-3.5-7B", "size": "4.0GB", "description": "OpenChat's latest version" },
+  { "name": "Dolphin-2.2.1-7B", "size": "4.1GB", "description": "Eric Hartford's uncensored model" },
+  { "name": "PlatYi-7B", "size": "4.0GB", "description": "01.AI's chat-optimized model" },
+  { "name": "TinyLlama-1.1B-Chat", "size": "1.1GB", "description": "Compact conversational model" },
+  { "name": "DeepSeek-LLM-7B", "size": "4.2GB", "description": "DeepSeek's language model" }
+
+
         "code": [
-            {"name": "CodeLlama-7B-Instruct", "size": "3.8GB", "description": "Instruction-tuned CodeLlama", "parameters": "7B", "context": "16K"},
-            {"name": "WizardCoder-7B", "size": "4.0GB", "description": "Enhanced coding abilities", "parameters": "7B", "context": "16K"},
-            {"name": "Phind-CodeLlama-34B-v2", "size": "4.2GB", "description": "4-bit quantized version", "parameters": "34B", "context": "16K"}
-        ],
-        "reasoning": [
-            {"name": "WizardMath-7B", "size": "4.0GB", "description": "Mathematical reasoning", "parameters": "7B", "context": "2K"},
-            {"name": "MetaMath-7B", "size": "3.9GB", "description": "Math problem solving", "parameters": "7B", "context": "2K"}
+            
+  { "name": "CodeLlama-7B-Instruct", "size": "3.8GB", "description": "Instruction-tuned code specialist" },
+  { "name": "WizardCoder-7B", "size": "4.0GB", "description": "Enhanced coding capabilities" },
+  { "name": "Phind-CodeLlama-7B-v2", "size": "3.9GB", "description": "Code search optimized model" },
+  { "name": "Magicoder-7B", "size": "4.0GB", "description": "OSS-Instruct trained code model" },
+  { "name": "DeepSeek-Coder-7B", "size": "3.9GB", "description": "DeepSeek's coding specialist" },
+  { "name": "WizardCoder-Python-7B", "size": "4.0GB", "description": "Python-specialized coding model" },
+  { "name": "StarCoder-7B", "size": "4.0GB", "description": "BigCode's 7B programming model" },
+  { "name": "CodeT5p-7B", "size": "4.1GB", "description": "Salesforce's code understanding" },
+  { "name": "InstructCodeT5p-7B", "size": "4.2GB", "description": "Instruction-tuned CodeT5p" },
+  { "name": "CodeGen2-7B", "size": "4.0GB", "description": "Salesforce's improved code model" },
+  { "name": "SantaCoder-7B", "size": "4.1GB", "description": "BigCode's multilingual coder" },
+  { "name": "Replit-Code-7B", "size": "4.0GB", "description": "Replit's code completion model" },
+  { "name": "Code-Alpaca-7B", "size": "3.9GB", "description": "Stanford's code instruction model" },
+  { "name": "UnixCoder-7B", "size": "4.0GB", "description": "Microsoft's large code model" }
+]
+  "chat": [
+       
+  { "name": "Vicuna-7B-v1.5", "size": "3.9GB", "description": "Enhanced conversational model" },
+  { "name": "ChatGLM2-6B", "size": "3.7GB", "description": "Tsinghua's bilingual chat model" },
+  { "name": "Baize-7B", "size": "4.0GB", "description": "Self-chat trained model" },
+  { "name": "OpenBuddy-7B", "size": "4.0GB", "description": "Cross-lingual AI assistant" },
+  { "name": "Koala-7B", "size": "3.9GB", "description": "UC Berkeley's dialogue model" },
+  { "name": "GPT4All-7B", "size": "4.0GB", "description": "Nomic AI's local chat model" },
+  { "name": "Wizard-Vicuna-7B", "size": "4.1GB", "description": "Combined instruction model" },
+  { "name": "Manticore-7B", "size": "4.0GB", "description": "Multi-domain chat model" },
+  { "name": "Airoboros-7B", "size": "4.1GB", "description": "Context-aware chat model" },
+  { "name": "Samantha-1.2-7B", "size": "4.0GB", "description": "Empathetic AI assistant" }
+
+     
+ 
+]
+ "reasoning": [
+  { "name": "MetaMath-7B", "size": "3.9GB", "description": "Mathematical problem solving" },
+  { "name": "Abel-7B", "size": "4.0GB", "description": "Advanced reasoning capabilities" },
+  { "name": "WizardMath-7B-V1.1", "size": "4.0GB", "description": "Enhanced math reasoning" },
+  { "name": "MAmmoTH-7B", "size": "4.1GB", "description": "Mathematical reasoning model" },
+  { "name": "Orca-2-7B", "size": "4.2GB", "description": "Microsoft's reasoning model" },
+  { "name": "OpenOrca-7B", "size": "4.0GB", "description": "Open-source Orca variant" }
+        ]
+ "multilingual": [
+
+  { "name": "Qwen-7B", "size": "4.0GB", "description": "Alibaba's multilingual foundation" },
+  { "name": "Baichuan2-7B", "size": "4.1GB", "description": "Chinese-English bilingual" },
+  { "name": "InternLM-7B", "size": "4.0GB", "description": "Shanghai AI Lab multilingual" },
+  { "name": "Chinese-LLaMA-2-7B", "size": "4.0GB", "description": "Chinese-optimized Llama" },
+  { "name": "Vigogne-7B", "size": "4.1GB", "description": "French instruction model" }
+
+
         ]
     },
     "good": {  # 9-16GB
         "general": [
-            {"name": "Llama-2-13B-Chat", "size": "7.3GB", "description": "Larger Llama variant", "parameters": "13B", "context": "4K"},
-            {"name": "Vicuna-13B-v1.5", "size": "7.2GB", "description": "Enhanced Vicuna", "parameters": "13B", "context": "16K"},
-            {"name": "OpenChat-3.5", "size": "7.1GB", "description": "High-quality chat model", "parameters": "7B", "context": "8K"},
-            {"name": "Nous-Hermes-2-Mixtral-8x7B-DPO", "size": "12.9GB", "description": "4-bit quantized MoE", "parameters": "47B", "context": "32K"}
+            
+  { "name": "Llama-2-13B-Chat", "size": "7.3GB", "description": "Larger Llama variant (4-bit)" },
+  { "name": "Vicuna-13B-v1.5", "size": "7.2GB", "description": "Enhanced large chat model" },
+  { "name": "OpenChat-3.5-13B", "size": "7.1GB", "description": "High-quality large chat model" },
+  { "name": "Qwen-14B-Chat", "size": "7.8GB", "description": "Alibaba's advanced model" },
+  { "name": "Baichuan2-13B-Chat", "size": "7.5GB", "description": "Large Chinese language model" },
+  { "name": "Yi-34B-Chat (8-bit)", "size": "19.5GB", "description": "01.AI's flagship model" },
+  { "name": "Nous-Hermes-13B", "size": "7.3GB", "description": "NousResearch's large assistant" },
+  { "name": "WizardLM-13B", "size": "7.2GB", "description": "Microsoft's instruction model" },
+  { "name": "Alpaca-13B", "size": "7.0GB", "description": "Stanford's large instruction model" },
+  { "name": "Llama-2-13B", "size": "7.0GB", "description": "Meta's 13B foundation model" },
+  { "name": "MPT-30B", "size": "15.0GB", "description": "MosaicML's large transformer" },
+  { "name": "Falcon-40B (8-bit)", "size": "20.0GB", "description": "TII's large instruction model" },
+  { "name": "Guanaco-13B", "size": "7.1GB", "description": "QLoRA fine-tuned model" },
+  { "name": "Orca-13B", "size": "7.4GB", "description": "Microsoft's reasoning model" },
+  { "name": "Platypus-13B", "size": "7.2GB", "description": "Fine-tuned Llama variant" },
+  { "name": "WizardLM-13B-V1.2", "size": "7.3GB", "description": "Improved WizardLM" },
+  { "name": "Nous-Hermes-2-13B", "size": "7.4GB", "description": "Enhanced Hermes model" },
+  { "name": "OpenOrca-13B", "size": "7.2GB", "description": "Open-source Orca recreation" },
+  { "name": "Airoboros-13B", "size": "7.3GB", "description": "Context-aware large model" },
+  { "name": "MythoMax-13B", "size": "7.2GB", "description": "Roleplay-optimized model" }
+
+
         ],
         "code": [
-            {"name": "CodeLlama-13B-Instruct", "size": "7.3GB", "description": "Larger code model", "parameters": "13B", "context": "16K"},
-            {"name": "WizardCoder-15B", "size": "8.2GB", "description": "Advanced coding", "parameters": "15B", "context": "16K"},
-            {"name": "StarCoder-15B", "size": "8.5GB", "description": "Large code model", "parameters": "15B", "context": "8K"}
+            
+  { "name": "CodeLlama-13B-Instruct", "size": "7.3GB", "description": "Large code generation model" },
+  { "name": "WizardCoder-15B", "size": "8.2GB", "description": "Advanced coding assistant" },
+  { "name": "StarCoder-15B", "size": "8.5GB", "description": "Large programming model" },
+  { "name": "CodeT5p-16B", "size": "8.8GB", "description": "Salesforce's large code model" },
+  { "name": "Phind-CodeLlama-34B (8-bit)", "size": "19.0GB", "description": "Large code search model" },
+  { "name": "DeepSeek-Coder-33B (8-bit)", "size": "18.5GB", "description": "Large coding specialist" },
+  { "name": "CodeLlama-13B-Python", "size": "7.4GB", "description": "Python-specialized CodeLlama" },
+  { "name": "WizardCoder-Python-13B", "size": "7.3GB", "description": "Python coding wizard" },
+  { "name": "InstructCodeT5p-16B", "size": "8.9GB", "description": "Large instruction code model" },
+  { "name": "CodeGen2-16B", "size": "8.7GB", "description": "Salesforce's large code model" }
+
+
         ],
         "multimodal": [
-            {"name": "LLaVA-7B", "size": "7.0GB", "description": "Vision + language", "parameters": "7B", "context": "2K"},
-            {"name": "MiniGPT-4-7B", "size": "6.8GB", "description": "Multimodal chat", "parameters": "7B", "context": "2K"},
-            {"name": "Instructblip-7B", "size": "7.2GB", "description": "Instruction-tuned VLM", "parameters": "7B", "context": "2K"}
+           
+  { "name": "LLaVA-13B", "size": "7.5GB", "description": "Large vision-language model" },
+  { "name": "MiniGPT-4-13B", "size": "7.2GB", "description": "Multimodal conversational AI" },
+  { "name": "InstructBLIP-13B", "size": "7.8GB", "description": "Vision-language instruction model" },
+  { "name": "BLIP-2-FlanT5-XL", "size": "4.8GB", "description": "Salesforce's vision-language model" },
+  { "name": "Flamingo-9B", "size": "9.0GB", "description": "DeepMind's few-shot learning model" },
+  { "name": "LLaVA-1.5-13B", "size": "7.6GB", "description": "Improved LLaVA model" },
+  { "name": "Otter-13B", "size": "7.4GB", "description": "Multi-modal instruction tuned" },
+  { "name": "mPLUG-Owl-14B", "size": "8.0GB", "description": "Alibaba's multimodal model" },
+  { "name": "InternLM-XComposer-7B", "size": "7.0GB", "description": "Vision-language composition" },
+  { "name": "Qwen-VL-7B", "size": "7.2GB", "description": "Qwen vision-language model" }
+
+ 
         ],
         "reasoning": [
-            {"name": "WizardMath-13B", "size": "7.3GB", "description": "Advanced math", "parameters": "13B", "context": "2K"},
-            {"name": "Orca-2-13B", "size": "7.4GB", "description": "Microsoft reasoning", "parameters": "13B", "context": "4K"}
+            
+  { "name": "WizardMath-13B", "size": "7.3GB", "description": "Advanced mathematical reasoning" },
+  { "name": "Orca-2-13B", "size": "7.4GB", "description": "Microsoft's reasoning specialist" },
+  { "name": "MetaMath-13B", "size": "7.2GB", "description": "Mathematical problem solver" },
+  { "name": "MAmmoTH-13B", "size": "7.3GB", "description": "Large mathematical reasoning model" },
+  { "name": "Abel-13B", "size": "7.4GB", "description": "Advanced reasoning capabilities" },
+  { "name": "Goat-13B", "size": "7.2GB", "description": "Arithmetic reasoning specialist" },
+  { "name": "OpenOrca-Platypus-13B", "size": "7.3GB", "description": "Combined reasoning model" }
+
+        ]
+         "multilingual": [
+      { name: "Qwen-14B", size: "7.8GB", description: "Alibaba's large multilingual" },
+      { name: "Baichuan2-13B", size: "7.5GB", description: "Large Chinese-English model" },
+      { name: "InternLM-20B", size: "11.0GB", description: "Shanghai AI Lab's large model" },
+      { name: "Chinese-Alpaca-Plus-13B", size: "7.4GB", description: "Enhanced Chinese model" },
+      { name: "Polyglot-Ko-13B", size: "7.3GB", description: "Large Korean model" }
+
+
         ]
     },
     "high": {  # 17-32GB
         "general": [
-            {"name": "Mixtral-8x7B-Instruct-v0.1", "size": "26.9GB", "description": "Mixture of experts", "parameters": "47B", "context": "32K"},
-            {"name": "Llama-2-70B-Chat", "size": "38.0GB", "description": "8-bit quantized", "parameters": "70B", "context": "4K"},
-            {"name": "Yi-34B-Chat", "size": "19.5GB", "description": "01.AI's large model", "parameters": "34B", "context": "200K"},
-            {"name": "Nous-Hermes-2-Yi-34B", "size": "19.2GB", "description": "Enhanced Yi variant", "parameters": "34B", "context": "200K"}
+            
+  { "name": "Mixtral-8x7B-Instruct-v0.1", "size": "26.9GB", "description": "Mixture of experts model (4-bit)" },
+  { "name": "Llama-2-70B-Chat (8-bit)", "size": "38.0GB", "description": "Large language model" },
+  { "name": "Yi-34B-Chat", "size": "19.5GB", "description": "01.AI's flagship model" },
+  { "name": "Qwen-72B (4-bit)", "size": "36.0GB", "description": "Alibaba's largest model" },
+  { "name": "DeepSeek-67B", "size": "35.0GB", "description": "Advanced reasoning model" },
+  { "name": "Nous-Hermes-2-Mixtral-8x7B", "size": "26.9GB", "description": "NousResearch's MoE model" },
+  { "name": "Solar-10.7B", "size": "10.7GB", "description": "Upstage's efficient model" },
+  { "name": "Dolphin-2.5-Mixtral-8x7B", "size": "26.9GB", "description": "Uncensored Mixtral variant" },
+  { "name": "Llama-2-70B", "size": "35.0GB", "description": "Meta's flagship model (8-bit)" },
+  { "name": "Falcon-40B", "size": "20.0GB", "description": "TII's large model" },
+  { "name": "MPT-30B", "size": "15.0GB", "description": "MosaicML's 30B model" },
+  { "name": "Nous-Hermes-2-Yi-34B", "size": "19.6GB", "description": "Enhanced Yi model" },
+  { "name": "OpenHermes-2.5-Mistral-7B", "size": "4.1GB", "description": "Teknium's Mistral variant" },
+  { "name": "Starling-LM-7B-alpha", "size": "4.2GB", "description": "Berkeley's RLAIF model" },
+  { "name": "NeuralBeagle-14B", "size": "8.0GB", "description": "MLP KAT merged model" },
+  { "name": "Goliath-120B (4-bit)", "size": "60.0GB", "description": "Large merged model" },
+  { "name": "Xwin-LM-70B (8-bit)", "size": "38.5GB", "description": "Xwin team's large model" },
+  { "name": "Airoboros-L2-70B (8-bit)", "size": "38.0GB", "description": "Large context model" }
+
+
         ],
         "code": [
-            {"name": "CodeLlama-34B-Instruct", "size": "19.0GB", "description": "Large code specialist", "parameters": "34B", "context": "16K"},
-            {"name": "DeepSeek-Coder-33B", "size": "18.5GB", "description": "DeepSeek's coder", "parameters": "33B", "context": "16K"},
-            {"name": "WizardCoder-34B", "size": "19.2GB", "description": "Enterprise coding", "parameters": "34B", "context": "16K"}
+            
+  { "name": "CodeLlama-34B-Instruct", "size": "19.0GB", "description": "Large specialized coder" },
+  { "name": "DeepSeek-Coder-33B", "size": "18.5GB", "description": "Advanced code generation" },
+  { "name": "WizardCoder-34B", "size": "19.2GB", "description": "Enterprise-grade coding" },
+  { "name": "StarCoder2-15B", "size": "8.5GB", "description": "Next-gen programming model" },
+  { "name": "Phind-CodeLlama-34B", "size": "19.0GB", "description": "Code search specialized model" },
+  { "name": "Magicoder-34B", "size": "19.1GB", "description": "Large OSS-Instruct model" },
+  { "name": "CodeLlama-34B-Python", "size": "19.1GB", "description": "Python-specialized large model" },
+  { "name": "WizardCoder-Python-34B", "size": "19.2GB", "description": "Large Python specialist" },
+  { "name": "StarCoder-15.5B", "size": "8.8GB", "description": "Enhanced StarCoder" },
+  { "name": "Code-Alpaca-34B", "size": "18.9GB", "description": "Large code instruction model" }
+
+
         ],
+         "chat": [
+            
+   
+  { "name": "Vicuna-33B", "size": "18.5GB", "description": "Large conversational model" },
+  { "name": "Guanaco-65B (4-bit)", "size": "33.0GB", "description": "Large instruction-tuned model" },
+  { "name": "Alpaca-30B", "size": "18.0GB", "description": "Large Stanford model" },
+  { "name": "OpenBuddy-34B", "size": "19.0GB", "description": "Large cross-lingual assistant" },
+  { "name": "WizardLM-30B", "size": "17.0GB", "description": "Large instruction model" },
+  { "name": "Nous-Hermes-Llama2-70B (8-bit)", "size": "38.2GB", "description": "Large Hermes variant" },
+  { "name": "Airoboros-65B (4-bit)", "size": "33.5GB", "description": "Large context chat model" },
+  { "name": "MythoMax-L2-13B", "size": "7.4GB", "description": "Roleplay optimized" }
+],
         "reasoning": [
-            {"name": "WizardMath-70B", "size": "38.5GB", "description": "8-bit quantized math", "parameters": "70B", "context": "2K"},
-            {"name": "MetaMath-70B", "size": "38.0GB", "description": "8-bit math reasoning", "parameters": "70B", "context": "2K"}
-        ]
-    },
+             
+    { "name": "WizardMath-70B (8-bit)", "size": "38.5GB", "description": "Premier math reasoning" },
+    { "name": "MetaMath-70B (8-bit)", "size": "38.0GB", "description": "Advanced mathematical AI" },
+    { "name": "Goat-70B (8-bit)", "size": "35.0GB", "description": "Arithmetic reasoning specialist" },
+    { "name": "MAmmoTH-70B (8-bit)", "size": "38.2GB", "description": "Large mathematical model" },
+    { "name": "Orca-2-13B", "size": "7.4GB", "description": "Microsoft's reasoning model" },
+    { "name": "Abel-70B (8-bit)", "size": "38.1GB", "description": "Large reasoning model" }
+  ]
+}
+            },
     "ultra_high": {  # >32GB
         "general": [
             {"name": "Llama-2-70B", "size": "130GB", "description": "Full precision", "parameters": "70B", "context": "4K"},