Q3 ThinkerMOE BaseTest0 is an open-source language model by SvalTek. Features: 7b LLM, VRAM: 14GB, Context: 40K, License: apache-2.0, Quantized.
| LLM Name | Q3 ThinkerMOE BaseTest0 |
| Repository π€ | https://huggingface.co/SvalTek/Q3-ThinkerMOE-BaseTest0 |
| Model Size | 7b |
| Required VRAM | 14 GB |
| Updated | 2026-11-18 |
| Maintainer | SvalTek |
| Model Type | qwen3_moe |
| Model Files | |
| Supported Languages | en |
| Quantization Type | q3 |
| Model Architecture | Qwen3MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 40960 |
| Model Max Length | 40960 |
| Transformers Version | 5.3.0 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 151936 |
| Torch Data Type | bfloat16 |
| Errors | replace |
Model |
Likes |
Downloads |
VRAM |
|---|---|---|---|
| Q3 ThinkerMOE Base0 | 0 | 114 | 14 GB |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Q3 ThinkerMOE TermTest0 | 40K / 14 GB | 165 | 0 |
| Q3 ThinkerMOE Base0 | 40K / 14 GB | 114 | 0 |
| ....2B Shining Madness Uncensored | 40K / 14 GB | 65 | 3 |