LLM Name | Qwen3 30B A3B Instruct 2507 MLX 4bit |
Repository ๐ค | https://huggingface.co/lmstudio-community/Qwen3-30B-A3B-Instruct-2507-MLX-4bit |
Base Model(s) | |
Model Size | 30b |
Required VRAM | 17.2 GB |
Updated | 2025-07-31 |
Maintainer | lmstudio-community |
Model Type | qwen3_moe |
Instruction-Based | Yes |
Model Files | |
Quantization Type | 4bit |
Model Architecture | Qwen3MoeForCausalLM |
License | apache-2.0 |
Context Length | 262144 |
Model Max Length | 262144 |
Transformers Version | 4.51.0 |
Tokenizer Class | Qwen2Tokenizer |
Padding Token | <|endoftext|> |
Vocabulary Size | 151936 |
Torch Data Type | bfloat16 |
Errors | replace |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
---|---|---|---|
...30B A3B Instruct 2507 MLX 8bit | 256K / 32.5 GB | 25323 | 2 |
...en3 30B A3B Instruct 2507 4bit | 256K / 61.2 GB | 72 | 2 |
Q3 30B A3B Designant | 32K / 61.1 GB | 25 | 13 |
Qwen3 30B A3B Instruct 2507 | 256K / 61.1 GB | 1489 | 259 |
...wen3 30B A3B Instruct 2507 FP8 | 256K / 31.2 GB | 1947 | 43 |
Qwen3 30B A3B Instruct 2507 | 256K / 61.1 GB | 87 | 4 |
...wen3 30B A3B Instruct 2507 FP8 | 256K / 31.2 GB | 23 | 4 |
๐ Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! ๐