Skip to content

armgit / Soup

Public Imported from GitHub
0

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

3.4 KB
Raw Blame
{
  "schema_version": 1,
  "tag": "b5270",
  "commit": "3bf785f3efa89ed28294fbf73054558a2b034bfb",
  "source_url": "https://github.com/ggml-org/llama.cpp/blob/3bf785f3efa89ed28294fbf73054558a2b034bfb/convert_hf_to_gguf.py",
  "source_sha256": "b69d3b6a51636b580966bf931088fa095d88cd3f13bc91083eb85b29c23eb05e",
  "source_git_blob": "838999531e58041976c8e612bc97b382874bf227",
  "architectures": {
    "TEXT": [
      "ArcticForCausalLM",
      "BaiChuanForCausalLM",
      "BaichuanForCausalLM",
      "BailingMoeForCausalLM",
      "BertForMaskedLM",
      "BertModel",
      "BitnetForCausalLM",
      "BloomForCausalLM",
      "BloomModel",
      "CamembertModel",
      "ChameleonForCausalLM",
      "ChameleonForConditionalGeneration",
      "ChatGLMForConditionalGeneration",
      "ChatGLMModel",
      "CodeShellForCausalLM",
      "Cohere2ForCausalLM",
      "CohereForCausalLM",
      "DbrxForCausalLM",
      "DeciLMForCausalLM",
      "DeepseekForCausalLM",
      "DeepseekV2ForCausalLM",
      "DeepseekV3ForCausalLM",
      "ExaoneForCausalLM",
      "FalconForCausalLM",
      "FalconMambaForCausalLM",
      "GPT2LMHeadModel",
      "GPTBigCodeForCausalLM",
      "GPTNeoXForCausalLM",
      "GPTRefactForCausalLM",
      "Gemma2ForCausalLM",
      "Gemma3ForCausalLM",
      "Gemma3ForConditionalGeneration",
      "GemmaForCausalLM",
      "Glm4ForCausalLM",
      "GlmForCausalLM",
      "GraniteForCausalLM",
      "GraniteMoeForCausalLM",
      "GrokForCausalLM",
      "InternLM2ForCausalLM",
      "InternLM3ForCausalLM",
      "JAISLMHeadModel",
      "JinaBertForMaskedLM",
      "JinaBertModel",
      "LLaMAForCausalLM",
      "Llama4ForConditionalGeneration",
      "LlamaForCausalLM",
      "LlavaForConditionalGeneration",
      "LlavaStableLMEpochForCausalLM",
      "MPTForCausalLM",
      "MT5ForConditionalGeneration",
      "MambaForCausalLM",
      "MambaLMHeadModel",
      "MiniCPM3ForCausalLM",
      "MiniCPMForCausalLM",
      "Mistral3ForConditionalGeneration",
      "MistralForCausalLM",
      "MixtralForCausalLM",
      "NemotronForCausalLM",
      "NomicBertModel",
      "OLMoForCausalLM",
      "Olmo2ForCausalLM",
      "OlmoForCausalLM",
      "OlmoeForCausalLM",
      "OpenELMForCausalLM",
      "OrionForCausalLM",
      "PLMForCausalLM",
      "Phi3ForCausalLM",
      "PhiForCausalLM",
      "PhiMoEForCausalLM",
      "PlamoForCausalLM",
      "QWenLMHeadModel",
      "Qwen2ForCausalLM",
      "Qwen2Model",
      "Qwen2MoeForCausalLM",
      "Qwen2VLForConditionalGeneration",
      "Qwen2_5_VLForConditionalGeneration",
      "Qwen3ForCausalLM",
      "Qwen3MoeForCausalLM",
      "RWForCausalLM",
      "RWKV6Qwen2ForCausalLM",
      "RWKV7ForCausalLM",
      "RobertaModel",
      "Rwkv6ForCausalLM",
      "Rwkv7ForCausalLM",
      "RwkvHybridForCausalLM",
      "StableLMEpochForCausalLM",
      "StableLmForCausalLM",
      "Starcoder2ForCausalLM",
      "T5EncoderModel",
      "T5ForConditionalGeneration",
      "T5WithLMHeadModel",
      "UMT5ForConditionalGeneration",
      "VLlama3ForCausalLM",
      "WavTokenizerDec",
      "XLMRobertaForSequenceClassification",
      "XLMRobertaModel",
      "XverseForCausalLM"
    ],
    "VISION": [
      "Gemma3ForConditionalGeneration",
      "Idefics3ForConditionalGeneration",
      "LlavaForConditionalGeneration",
      "Mistral3ForConditionalGeneration",
      "Qwen2VLForConditionalGeneration",
      "Qwen2_5_VLForConditionalGeneration",
      "SmolVLMForConditionalGeneration"
    ]
  }
}