LLM EXPLORER 60,384 MODELS INDEXED

Qwen3.6 Whittle 25B A3B by logic65

By logic65 · 611 downloads
HF Repo ↗ Copy URL

Qwen3.6 Whittle 25B A3B is an open-source language model by logic65. Features: 25.1b LLM, VRAM: 49.9GB, Context: 256K, License: apache-2.0, MoE, LLM Explorer Score: 0.33.

Base model:quantized:qwen/qwen... Base model:qwen/qwen3.6-35b-a3...   Conversational   Distillation   Endpoints compatible   Expert-pruning   Gguf   Moe   Qwen3.6   Qwen3 5 moe text   Region:us   Safetensors   Sharded   Tensorflow   Whittle
Model Card on HF 🤗: https://huggingface.co/logic65/Qwen3.6-Whittle-25B-A3B 

Qwen3.6 Whittle 25B A3B Parameters and Internals

LLM NameQwen3.6 Whittle 25B A3B
Repository 🤗https://huggingface.co/logic65/Qwen3.6-Whittle-25B-A3B 
Base Model(s)  Qwen/Qwen3.6-35B-A3B   Qwen/Qwen3.6-35B-A3B
Model Size25.1b
Required VRAM49.9 GB
Updated2026-09-04
Maintainerlogic65
Model Typeqwen3_5_moe_text
Model Files  3.6 GB: 1-of-15   3.4 GB: 2-of-15   3.4 GB: 3-of-15   3.4 GB: 4-of-15   3.4 GB: 5-of-15   3.4 GB: 6-of-15   3.4 GB: 7-of-15   3.4 GB: 8-of-15   3.4 GB: 9-of-15   3.4 GB: 10-of-15   3.4 GB: 11-of-15   3.4 GB: 12-of-15   3.4 GB: 13-of-15   3.4 GB: 14-of-15   2.1 GB: 15-of-15
Model ArchitectureQwen3_5MoeForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Transformers Version4.57.1
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size248320
Errorsreplace

Best Alternatives to Qwen3.6 Whittle 25B A3B

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3.5 24.5B Reapped V1256K / 50.1 GB1183
...n3.6 35B A3B Heretic REAP 0.30256K / 50.2 GB664