Ministral 14B
Trinity Nano Preview
- Context
- 33Ktokens
- Released
- 1y ago
- Size
- 8Bparameters
Qwen3-8B is an open-weight model from Alibaba, one of six dense models in the Qwen3 series released under the Apache 2.0 license. Its launch headline is a hybrid design that switches between extended step-by-step reasoning and quick direct answers, and the company claims support for 119 languages and dialects. Alibaba says it optimized the Qwen3 models for coding and agentic capabilities.
- Language model
- Open weights · 8B
- Reasoning
- Tools
How it compares
Size class
Small8BAn open-weight model of 5 to 25 billion parameters, the size that runs on a single consumer GPU.
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/qwen3-8b
- Source family
- qwen
- Last updated
- 2025-04-29
- Model context
- 33K
- Model output limit
- 33K
- Open weights
- Yes
- File attachments
- No
- Reasoning
- Yes
- Tool calling
- Yes
- Temperature control
- Yes
Downloadable weights
Complete source record (JSON)
{
"id": "alibaba/qwen3-8b",
"name": "Qwen3 8B",
"description": "Dense 8B Qwen3 model with switchable thinking for chat, reasoning, and tool use",
"family": "qwen",
"last_updated": "2025-04-29",
"attachment": false,
"reasoning": true,
"tool_call": true,
"temperature": true,
"modalities": {
"input": [
"text"
],
"output": [
"text"
]
},
"limit": {
"context": 32768,
"output": 32768
},
"release_date": "2025-04-29",
"open_weights": true,
"license": "Apache-2.0",
"weights": [
{
"label": "Hugging Face",
"url": "https://huggingface.co/Qwen/Qwen3-8B"
}
]
}Frequently asked questions
Running locally
Qwen3 8B needs an estimated 6.1 GB of VRAM at Q4, 10.3 GB at Q8 and 18.9 GB at FP16. These figures are estimates, with weights plus about 15% and 0.5 GB for a short context.
Yes. A 24 GB GPU fits FP16 (estimated 18.9 GB), 16 GB and 12 GB GPUs fit Q8 (estimated 10.3 GB) and an 8 GB GPU fits Q4 (estimated 6.1 GB).
Yes, at Q4 the estimated 6.1 GB fits. Assuming macOS lets the GPU use about 75% of unified memory, the smallest common Mac memory size that fits Q4 is 16 GB.
Privacy and data
Yes. Qwen3 8B has open weights, so you can self-host it for full control of your data.
Capabilities and specs
Qwen3 8B has a context window of 33K tokens (32,768) and a maximum output of 33K tokens (32,768).
No. Qwen3 8B only accepts text input and cannot read images, audio, video or PDFs.
Yes, Qwen3 8B supports tool calling (function calling).
Yes, Qwen3 8B supports a reasoning mode.
Quality and benchmarks
Matched on model kind and release date (no comparable price), the closest alternatives are FastVLM 0.5B and FastVLM 1.5B by Apple and Sarvam-M by Sarvam AI.
FastVLM 0.5BQwen3 8B vs FastVLM 0.5BFastVLM 1.5BQwen3 8B vs FastVLM 1.5BSarvam-MQwen3 8B vs Sarvam-M
Versions and lineage
Qwen3 8B has a 33K token (32,768) context and was released April 29, 2025. Qwen3 235B-A22B has a 131K token (131,072) context and was released April 28, 2025.
Licensing and access
Qwen3 8B is open weights, released under the Apache-2.0 license.
Yes. Qwen3 8B uses the Apache 2.0 license, a permissive license that allows commercial use.
Basics
Qwen3 8B was made by Qwen and released on April 29, 2025.
Yes, Qwen3 8B has open weights.
Generated from Vertical Tabs data, updated November 10, 2026.
Discussion
No comments yet
Be the first to share your thoughts.