- Context
- 262Ktokens
- Released
- 8mo ago
- Size
- 0.8Bparameters
Smallest Qwen3.5 multimodal model for on-device chat, reasoning, and tool use
- Multimodal model
- Open weights · 0.8B
- Accepts images, video
- Reasoning
- Tools
How it compares
Size class
Tiny800MAn open-weight model under 5 billion parameters, small enough for phones, laptops and edge devices.
Open models of a similar size from other labs
Vs. your stack
Add the models you use to see how this one compares.
Sign inAdditional source evidence
- Source model
- alibaba/qwen3.5-0.8b
- Source family
- qwen
- Last updated
- 2026-03-02
- Model context
- 262K
- Model output limit
- 82K
- Open weights
- Yes
- File attachments
- Yes
- Reasoning
- Yes
- Tool calling
- Yes
- Temperature control
- Yes
Downloadable weights
Complete source record (JSON)
{
"id": "alibaba/qwen3.5-0.8b",
"name": "Qwen3.5 0.8B",
"description": "Smallest Qwen3.5 multimodal model for on-device chat, reasoning, and tool use",
"family": "qwen",
"last_updated": "2026-03-02",
"attachment": true,
"reasoning": true,
"tool_call": true,
"temperature": true,
"modalities": {
"input": [
"text",
"image",
"video"
],
"output": [
"text"
]
},
"limit": {
"context": 262144,
"output": 81920
},
"release_date": "2026-03-02",
"open_weights": true,
"license": "Apache-2.0",
"weights": [
{
"label": "Hugging Face",
"url": "https://huggingface.co/Qwen/Qwen3.5-0.8B"
}
]
}Frequently asked questions
Running locally
An estimated 1.1 GB at Q4, 1.5 GB at Q8 or 2.3 GB at FP16. These estimates come from the 0.8B parameter count, using weights plus about 15% and 0.5 GB for a short context.
Yes. Qwen3.5 0.8B fits at FP16 on a 24 GB, 16 GB, 12 GB or 8 GB GPU, with an estimated need of 2.3 GB.
Yes. At Q4 the estimated VRAM is 1.1 GB, and assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits is 8 GB.
Privacy and data
Yes. Qwen3.5 0.8B has open weights, so you can self-host it for full control of your data.
Capabilities and specs
The context window is 262K tokens (262,144) and the maximum output is 82K tokens (81,920).
Qwen3.5 0.8B reads text, images and video. Audio and PDF are not among its listed input modalities.
Yes, Qwen3.5 0.8B supports tool calling (function calling).
Yes, Qwen3.5 0.8B has a reasoning or thinking mode.
Quality and benchmarks
The closest alternatives, matched on model kind and release date with no comparable price, are Olmo Hybrid 7B Instruct and Olmo Hybrid 7B Think by Ai2 and Nemotron Cascade 2 30B A3B by NVIDIA.
Olmo Hybrid 7B InstructQwen3.5 0.8B vs Olmo Hybrid 7B InstructOlmo Hybrid 7B ThinkQwen3.5 0.8B vs Olmo Hybrid 7B ThinkNemotron Cascade 2 30B A3BQwen3.5 0.8B vs Nemotron Cascade 2 30B A3B
Versions and lineage
Both have a context window of 262K tokens (262,144) and were released on March 2, 2026. No other differences are listed in the facts.
Licensing and access
Qwen3.5 0.8B is open weights, released under the Apache-2.0 license.
Yes. Qwen3.5 0.8B uses the Apache 2.0 license, a permissive license that allows commercial use.
Basics
Qwen made Qwen3.5 0.8B and released it on March 2, 2026.
Yes, Qwen3.5 0.8B has open weights.
Generated from Vertical Tabs data, updated November 10, 2026.
Discussion
No comments yet
Be the first to share your thoughts.