DeepSeek V3.1 Terminus W4AFP8 is an open-source language model by tencent. Features: 349.2b LLM, VRAM: 366.7GB, Context: 160K, License: mit, LLM Explorer Score: 0.22.
| LLM Name | DeepSeek V3.1 Terminus W4AFP8 |
| Repository 🤗 | https://huggingface.co/tencent/DeepSeek-V3.1-Terminus-W4AFP8 |
| Base Model(s) | |
| Model Size | 349.2b |
| Required VRAM | 366.7 GB |
| Updated | 2026-08-08 |
| Maintainer | tencent |
| Model Type | deepseek_v3 |
| Model Files | |
| Model Architecture | DeepseekV3ForCausalLM |
| License | mit |
| Context Length | 163840 |
| Model Max Length | 163840 |
| Transformers Version | 4.44.2 |
| Tokenizer Class | LlamaTokenizerFast |
| Beginning of Sentence Token | <|begin▁of▁sentence|> |
| End of Sentence Token | <|end▁of▁sentence|> |
| Vocabulary Size | 129280 |
| Torch Data Type | bfloat16 |