| LLM Name | Qwen3 30B A3B AWQ |
| Repository ๐ค | https://huggingface.co/QuixiAI/Qwen3-30B-A3B-AWQ |
| Base Model(s) | |
| Model Size | 30b |
| Required VRAM | 16.8 GB |
| Updated | 2025-10-26 |
| Maintainer | QuixiAI |
| Model Type | qwen3_moe |
| Model Files | |
| AWQ Quantization | Yes |
| Quantization Type | awq |
| Model Architecture | Qwen3MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 40960 |
| Model Max Length | 40960 |
| Transformers Version | 4.51.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | float16 |
| Errors | replace |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| ...oder 30B A3B Instruct AWQ 4bit | 256K / 16.7 GB | 91021 | 17 |
| ...30B A3B Thinking 2507 AWQ 4bit | 256K / 16.7 GB | 67401 | 8 |
| ...30B A3B Instruct 2507 AWQ 4bit | 256K / 18.1 GB | 33039 | 14 |
| ...wen3 30B A3B Thinking 2507 AWQ | 256K / 16.7 GB | 4825 | 6 |
| ...wen3 30B A3B Instruct 2507 AWQ | 256K / 16.7 GB | 3531 | 6 |
| ...en3 Coder 30B A3B Instruct AWQ | 256K / 16.7 GB | 2464 | 3 |
| Qwen3 30B A3B AWQ | 40K / 16.8 GB | 4483 | 12 |
| ...30B A3B Instruct 2507 MLX 4bit | 256K / 17.2 GB | 89287 | 5 |
| ...30B A3B Instruct 2507 MLX 8bit | 256K / 32.5 GB | 73643 | 4 |
| ...en3 30B A3B Instruct 2507 4bit | 256K / 61.2 GB | 9813 | 8 |
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