Qwen3.5-2B

Home/Models/Qwen/Qwen3.5-2B

Qwen/Qwen3.5-2B

kind: original
quant: bf16 (16-bit)
family: qwen
modality: image-text-to-text
root model: Qwen/Qwen3.5-2B-Base
provider: Qwen
license: apache-2.0
language: —
context: 262,144
library: transformers
vRAM (GB): —
size on disk: —
downloads: 2,395,693
likes: 342
created: —
last modified: 2026-03-02T11:26:29+00:00 (2026)
huggingface: https://huggingface.co/Qwen/Qwen3.5-2B
tags: transformerssafetensorsqwen3_5image-text-to-textconversationalbase_model:Qwen/Qwen3.5-2B-Basebase_model:finetune:Qwen/Qwen3.5-2B-Baselicense:apache-2.0eval-resultsendpoints_compatibledeploy:azureregion:usdeploy:sagemaker

no scores extracted for this variant.

title	author	status	type
Add IFStruct v1.0 evaluation result	SaylorTwift	open	disc
Qwen3.5-2B Brain Atlas — The Editable One	juiceb0xc0de	open	disc
Add EvalEval community eval results	EvalEvalBot	open	disc
Add MMMU-Pro evaluation result	SaylorTwift	open	disc
Add ParseBench evaluation results	boyang-runllama	open	disc
fix chat template to avoid empty historical `<think>` blocks	latent-variable	open	disc
FP8 version?	drguolai	open	disc
Add ScreenSpot-Pro evaluation result	merve	open	disc
Title: Fix enable_thinking default behavior in Qwen3.5-2B chat template to match Qwen3.5 other size.	geray399	closed	disc
Installation Video and Testing - Step by Step	fahdmirzac	open	disc
Add MMLU-Pro evaluation result	SaylorTwift	open	disc