LLM EXPLORER 67,627 MODELS INDEXED

Sharpen Qwen 3.5 9B A2.6B Preview by Takenoko12345678

By Takenoko12345678 · 156 downloads

Sharpen Qwen 3.5 9B A2.6B Preview is an open-source language model by Takenoko12345678. Features: 2.4b LLM, VRAM: 0.1GB, Context: 256K, License: apache-2.0, MoE, Quantized, LLM Explorer Score: 0.48, ELO: 1395.

Base model:quantized:qwen/qwen... Base model:qwen/qwen3.5-35b-a3...   Conversational   En   Endpoints compatible   Expert-pruning   Gguf   Mixture-of-experts   Moe   Preview   Pruning   Q4   Quantized   Qwen3 5 moe text   Region:us   Safetensors

Sharpen Qwen 3.5 9B A2.6B Preview Benchmarks

nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

Sharpen Qwen 3.5 9B A2.6B Preview Parameters and Internals

LLM NameSharpen Qwen 3.5 9B A2.6B Preview
Repository πŸ€—https://huggingface.co/Takenoko12345678/Sharpen-Qwen-3.5-9B-A2.6B-Preview 
Base Model(s)  Qwen/Qwen3.5-35B-A3B   Qwen/Qwen3.5-35B-A3B
Model Size2.4b
Required VRAM0.1 GB
Updated2026-11-16
MaintainerTakenoko12345678
Model Typeqwen3_5_moe_text
Model Files  11.4 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB   0.1 GB
Supported Languagesen
GGUF QuantizationYes
Quantization Typegguf|q4|q4_k
Model ArchitectureQwen3_5MoeForCausalLM
Licenseapache-2.0
Context Length262144
Model Max Length262144
Vocabulary Size248320
Torch Data Typebfloat16