sealad886/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill-mlx

GGUF to MLX Conversion

• Generated: 2026-02-12T23:25:33.024611+00:00
• Source repo: TeichAI/GLM-4.7-Flash-Claude-Opus-4.5-High-Reasoning-Distill-GGUF
• Source filename: glm-4.7-flash-claude-4.5-opus.bf16.gguf
• Source revision: main
• Tensor count: 844

Weights

• weights-00001.safetensors
• weights-00002.safetensors
• weights-00003.safetensors
• weights-00004.safetensors
• weights-00005.safetensors
• weights-00006.safetensors
• weights-00007.safetensors
• weights-00008.safetensors
• weights-00009.safetensors
• weights-00010.safetensors
• weights-00011.safetensors
• weights-00012.safetensors
• weights-00013.safetensors
• weights-00014.safetensors
• weights-00015.safetensors
• weights-00016.safetensors
• weights-00017.safetensors
• weights-00018.safetensors
• weights-00019.safetensors
• weights-00020.safetensors
• weights-00021.safetensors
• weights-00022.safetensors
• weights-00023.safetensors
• weights-00024.safetensors
• weights-00025.safetensors
• weights-00026.safetensors
• weights-00027.safetensors
• weights-00028.safetensors
• weights-00029.safetensors
• weights-00030.safetensors
• weights-00031.safetensors
• weights-00032.safetensors
• weights-00033.safetensors
• weights-00034.safetensors
• weights-00035.safetensors
• weights-00036.safetensors
• weights-00037.safetensors
• weights-00038.safetensors
• weights-00039.safetensors
• weights-00040.safetensors
• weights-00041.safetensors
• weights-00042.safetensors
• weights-00043.safetensors
• weights-00044.safetensors
• weights-00045.safetensors
• weights-00046.safetensors
• weights-00047.safetensors
• weights-00048.safetensors

Config

{
  "bos_token_id": 154822,
  "eos_token_id": 154820,
  "hidden_size": 2048,
  "intermediate_size": 10240,
  "max_position_embeddings": 202752,
  "model_type": "deepseek2",
  "moe_intermediate_size": 1536,
  "num_attention_heads": 20,
  "num_experts": 64,
  "num_experts_per_tok": 4,
  "num_experts_used": 1,
  "num_hidden_layers": 47,
  "num_key_value_heads": 1,
  "num_local_experts": 1,
  "pad_token_id": 154821,
  "rms_norm_eps": 9.999999747378752e-06,
  "rope_theta": 1000000.0,
  "unk_token_id": 154820,
  "vocab_size": 154880
}

Tokenizer

• Model: gpt2
• Pre-tokenizer: glm4

Generated files:

• tokenizer.json
• tokenizer_config.json
• special_tokens_map.json

Provenance

Converted with gguf_to_mlx.py using the gguf Python library and MLX safetensors output.