Qwen3.8 2.4T A95B MLX Reap60 3bit is an open-source language model by pipenetwork. Features: 124.9b LLM, VRAM: 12.4GB, Context: 256K, License: apache-2.0, MoE, Quantized, LLM Explorer Score: 0.34.
| LLM Name | Qwen3.8 2.4T A95B MLX Reap60 3bit |
| Repository 🤗 | https://huggingface.co/pipenetwork/Qwen3.8-2.4T-A95B-MLX-reap60-3bit |
| Base Model(s) | |
| Model Size | 124.9b |
| Required VRAM | 12.4 GB |
| Updated | 2026-08-14 |
| Maintainer | pipenetwork |
| Model Type | qwen3_5_moe_text |
| Model Files | |
| Quantization Type | 3bit |
| Model Architecture | Qwen3_5MoeForCausalLM |
| License | apache-2.0 |
| Context Length | 262144 |
| Model Max Length | 262144 |
| Transformers Version | 4.57.3 |
| Tokenizer Class | Qwen2Tokenizer |
| Padding Token | <|endoftext|> |
| Vocabulary Size | 248320 |
| Errors | replace |