{ "01-ai/Yi-1.5-6B-Chat": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 4, "num_layers": 32, "torch_dtype": null, "total_b_params": 6.0 }, "01-ai/Yi-1.5-9B-Chat": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 4, "num_layers": 48, "torch_dtype": null, "total_b_params": 9.0 }, "EleutherAI/gpt-j-6b": { "architectures": [ "GPTJForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": null, "model_type": "gptj", "num_attention_heads": 16, "num_key_value_heads": 16, "num_layers": 28, "torch_dtype": null, "total_b_params": 6.05 }, "EleutherAI/gpt-neox-20b": { "architectures": [ "GPTNeoXForCausalLM" ], "hidden_size": 6144, "max_position_embeddings": 2048, "model_type": "gpt_neox", "num_attention_heads": 64, "num_key_value_heads": 64, "num_layers": 44, "torch_dtype": null, "total_b_params": 20.0 }, "HuggingFaceTB/SmolLM2-1.7B-Instruct": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 2048, "max_position_embeddings": 8192, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 32, "num_layers": 24, "torch_dtype": null, "total_b_params": 1.7 }, "Qwen/Qwen2.5-0.5B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 896, "max_position_embeddings": 32768, "model_type": "qwen2", "num_attention_heads": 14, "num_key_value_heads": 2, "num_layers": 24, "torch_dtype": null, "total_b_params": 0.49 }, "Qwen/Qwen2.5-1.5B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 1536, "max_position_embeddings": 32768, "model_type": "qwen2", "num_attention_heads": 12, "num_key_value_heads": 2, "num_layers": 28, "torch_dtype": null, "total_b_params": 1.54 }, "Qwen/Qwen2.5-3B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 2048, "max_position_embeddings": 32768, "model_type": "qwen2", "num_attention_heads": 16, "num_key_value_heads": 2, "num_layers": 36, "torch_dtype": null, "total_b_params": 3.09 }, "Qwen/Qwen2.5-7B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 3584, "max_position_embeddings": 32768, "model_type": "qwen2", "num_attention_heads": 28, "num_key_value_heads": 4, "num_layers": 28, "torch_dtype": null, "total_b_params": 7.62 }, "Qwen/Qwen3-14B": { "architectures": [ "Qwen3ForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 40960, "model_type": "qwen3", "num_attention_heads": 40, "num_key_value_heads": 8, "num_layers": 40, "torch_dtype": null, "total_b_params": 14.0 }, "allenai/OLMo-2-1124-13B-Instruct": { "architectures": [ "Olmo2ForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 4096, "model_type": "olmo2", "num_attention_heads": 40, "num_key_value_heads": 40, "num_layers": 40, "torch_dtype": null, "total_b_params": 14.0 }, "allenai/OLMo-2-1124-7B-Instruct": { "architectures": [ "Olmo2ForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "olmo2", "num_attention_heads": 32, "num_key_value_heads": 32, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.0 }, "berkeley-nest/Starling-LM-7B-alpha": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 8192, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.0 }, "deepseek-ai/deepseek-llm-7b-chat": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 32, "num_layers": 30, "torch_dtype": null, "total_b_params": 7.0 }, "google/gemma-2-2b-it": { "architectures": [ "Gemma2ForCausalLM" ], "hidden_size": 2304, "max_position_embeddings": 8192, "model_type": "gemma2", "num_attention_heads": 8, "num_key_value_heads": 4, "num_layers": 26, "torch_dtype": null, "total_b_params": 2.61 }, "google/gemma-3-12b-it": { "architectures": [ "Gemma3ForConditionalGeneration" ], "hidden_size": 3840, "max_position_embeddings": null, "model_type": "gemma3", "num_attention_heads": 16, "num_key_value_heads": 8, "num_layers": 48, "torch_dtype": null, "total_b_params": 12.0 }, "google/gemma-3-27b-it": { "architectures": [ "Gemma3ForConditionalGeneration" ], "hidden_size": 5376, "max_position_embeddings": null, "model_type": "gemma3", "num_attention_heads": 32, "num_key_value_heads": 16, "num_layers": 62, "torch_dtype": null, "total_b_params": 27.0 }, "ibm-granite/granite-3.0-2b-instruct": { "architectures": [ "GraniteForCausalLM" ], "hidden_size": 2048, "max_position_embeddings": 4096, "model_type": "granite", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 40, "torch_dtype": null, "total_b_params": 2.0 }, "ibm-granite/granite-3.0-8b-instruct": { "architectures": [ "GraniteForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "granite", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 40, "torch_dtype": null, "total_b_params": 8.0 }, "internlm/internlm2-7b": { "architectures": [ "InternLM2ForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 32768, "model_type": "internlm2", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.0 }, "internlm/internlm2_5-20b-chat": { "architectures": [ "InternLM2ForCausalLM" ], "hidden_size": 6144, "max_position_embeddings": 32768, "model_type": "internlm2", "num_attention_heads": 48, "num_key_value_heads": 8, "num_layers": 48, "torch_dtype": null, "total_b_params": 20.0 }, "meta-llama/Llama-2-13b-chat-hf": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 40, "num_key_value_heads": 40, "num_layers": 40, "torch_dtype": null, "total_b_params": 13.0 }, "meta-llama/Llama-3.1-8B-Instruct": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 131072, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 8.03 }, "meta-llama/Llama-3.2-3B-Instruct": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 3072, "max_position_embeddings": 131072, "model_type": "llama", "num_attention_heads": 24, "num_key_value_heads": 8, "num_layers": 28, "torch_dtype": null, "total_b_params": 3.21 }, "microsoft/Phi-3-medium-4k-instruct": { "architectures": [ "Phi3ForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 4096, "model_type": "phi3", "num_attention_heads": 40, "num_key_value_heads": 10, "num_layers": 40, "torch_dtype": null, "total_b_params": 14.0 }, "microsoft/Phi-3-mini-4k-instruct": { "architectures": [ "Phi3ForCausalLM" ], "hidden_size": 3072, "max_position_embeddings": 4096, "model_type": "phi3", "num_attention_heads": 32, "num_key_value_heads": 32, "num_layers": 32, "torch_dtype": null, "total_b_params": 3.8 }, "microsoft/phi-2": { "architectures": [ "PhiForCausalLM" ], "hidden_size": 2560, "max_position_embeddings": 2048, "model_type": "phi", "num_attention_heads": 32, "num_key_value_heads": 32, "num_layers": 32, "torch_dtype": null, "total_b_params": 2.7 }, "microsoft/phi-4": { "architectures": [ "Phi3ForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 16384, "model_type": "phi3", "num_attention_heads": 40, "num_key_value_heads": 10, "num_layers": 40, "torch_dtype": null, "total_b_params": 14.7 }, "mistralai/Mistral-7B-Instruct-v0.3": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 32768, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.25 }, "mistralai/Mistral-Small-24B-Instruct-2501": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 32768, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 40, "torch_dtype": null, "total_b_params": 24.0 }, "mistralai/Mistral-Small-Instruct-2409": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 6144, "max_position_embeddings": 32768, "model_type": "mistral", "num_attention_heads": 48, "num_key_value_heads": 8, "num_layers": 56, "torch_dtype": null, "total_b_params": 24.0 }, "mlabonne/AlphaMonarch-7B": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 32768, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.0 }, "mlabonne/NeuralBeagle14-7B": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 32768, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 7.0 }, "nvidia/AceMath-1.5B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 1536, "max_position_embeddings": 4096, "model_type": "qwen2", "num_attention_heads": 12, "num_key_value_heads": 2, "num_layers": 28, "torch_dtype": null, "total_b_params": 1.5 }, "nvidia/AceMath-7B-Instruct": { "architectures": [ "Qwen2ForCausalLM" ], "hidden_size": 3584, "max_position_embeddings": 4096, "model_type": "qwen2", "num_attention_heads": 28, "num_key_value_heads": 4, "num_layers": 28, "torch_dtype": null, "total_b_params": 7.0 }, "nvidia/Llama-3.1-Minitron-4B-Width-Base": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 3072, "max_position_embeddings": 131072, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 5.0 }, "nvidia/Mistral-NeMo-Minitron-8B-Instruct": { "architectures": [ "MistralForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 8192, "model_type": "mistral", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 40, "torch_dtype": null, "total_b_params": 8.0 }, "nvidia/Nemotron-Mini-4B-Instruct": { "architectures": [ "NemotronForCausalLM" ], "hidden_size": 3072, "max_position_embeddings": 4096, "model_type": "nemotron", "num_attention_heads": 24, "num_key_value_heads": 8, "num_layers": 32, "torch_dtype": null, "total_b_params": 4.0 }, "openai-community/gpt2": { "architectures": [ "GPT2LMHeadModel" ], "hidden_size": 768, "max_position_embeddings": null, "model_type": "gpt2", "num_attention_heads": 12, "num_key_value_heads": 12, "num_layers": 12, "torch_dtype": null, "total_b_params": 0.124 }, "tiiuae/Falcon3-3B-Instruct": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 3072, "max_position_embeddings": 32768, "model_type": "llama", "num_attention_heads": 12, "num_key_value_heads": 4, "num_layers": 22, "torch_dtype": null, "total_b_params": 3.0 }, "upstage/SOLAR-10.7B-Instruct-v1.0": { "architectures": [ "LlamaForCausalLM" ], "hidden_size": 4096, "max_position_embeddings": 4096, "model_type": "llama", "num_attention_heads": 32, "num_key_value_heads": 8, "num_layers": 48, "torch_dtype": null, "total_b_params": 10.7 }, "upstage/solar-pro-preview-instruct": { "architectures": [ "SolarForCausalLM" ], "hidden_size": 5120, "max_position_embeddings": 4096, "model_type": "solar", "num_attention_heads": 40, "num_key_value_heads": 10, "num_layers": 64, "torch_dtype": null, "total_b_params": 22.1 } }