Llm Jp 4 33B Thinking is an open-source language model by llm-jp. Features: 33b LLM, VRAM: 66.8GB, Context: 64K, License: apache-2.0, LLM Explorer Score: 0.26.
| LLM Name | Llm Jp 4 33B Thinking |
| Repository π€ | https://huggingface.co/llm-jp/llm-jp-4-33b-thinking |
| Model Size | 33b |
| Required VRAM | 66.8 GB |
| Updated | 2026-08-19 |
| Maintainer | llm-jp |
| Model Type | llama |
| Model Files | |
| Supported Languages | en ja |
| Model Architecture | LlamaForCausalLM |
| License | apache-2.0 |
| Context Length | 65536 |
| Model Max Length | 65536 |
| Transformers Version | 4.51.0 |
| Tokenizer Class | PreTrainedTokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 196608 |
| Torch Data Type | bfloat16 |
Best Alternatives |
Context / RAM |
Downloads |
Likes |
|---|---|---|---|
| Nanbeige4.3 33B | 512K / 66.3 GB | 4785 | 655 |
| Nanbeige4.3 33B NVFP4 | 512K / 22.1 GB | 537 | 20 |
| Nanbeige4.3 33B OQ4e | 512K / 19.4 GB | 894 | 3 |
| Nanbeige4.3 33B NVFP4 | 512K / 4.7 GB | 870 | 0 |
| Nanbeige4.3 33B INT4 | 512K / 0 GB | 194 | 6 |
| Nanbeige4.3 33B FP8 | 512K / 35.6 GB | 203 | 4 |
| Nanbeige4.3 33B FP8 | 512K / 4.7 GB | 396 | 2 |
| Nanbeige4.3 33B Heretic NVFP4 | 512K / 22.1 GB | 266 | 3 |
| Nanbeige 4.3 33B Heretic | 512K / 65.9 GB | 471 | 0 |
| Nanbeige4.3 33B NVFP4 | 512K / 22.1 GB | 340 | 1 |