Qwen3.6 35B A3B
Tiny-active-parameter open MoE — a local-inference favorite.
Qwen3.6 35B A3B is an open-weight mixture-of-experts model with only 3B active parameters, making it fast on consumer GPUs and Apple Silicon while supporting image and video input.
- Provider
- Alibaba (Qwen)
- Type
- Language model
- Released
- Apr 26, 2026
- Context window
- 262K tokens
- Max output
- 66K tokens
- Price
- $0.10 input / $0.95 output per 1M tokens
- Input
- text, image, video
- Output
- text
- Open weights
- Yes
- License
- Apache 2.0
Best for
- Local & on-device
- High volume / low cost
- Real-time / low latency
- Vision
Strengths
- Runs fast locally
- Multimodal
More from Alibaba (Qwen)
- Qwen3.8 Max — Alibaba (Qwen). Alibaba’s 2.4T-parameter proprietary flagship with video understanding.
- Qwen3.8 2.4T A95B — Alibaba (Qwen). The open-weight variant of Qwen3.8 Max — the largest open model available.
- Qwen3.8 Flash — Alibaba (Qwen). Open multimodal reasoning model for coding, desktop interaction and long video.
- Qwen3-Embedding-8B — Alibaba (Qwen). Leading open-weight multilingual embedding model.
- Qwen-Image-3.0-Pro — Alibaba (Qwen). Alibaba’s flagship image model with strong text rendering.
- Qwen-Image-2.1 — Alibaba (Qwen). Highest-ranked open-weight image model in the arena.
- Wan 3.0 — Alibaba (Qwen). #1 in the Artificial Analysis text-to-video arena.
- Qwen-Audio-3.1-TTS-Plus — Alibaba (Qwen). Top-3 voice quality at a quarter of the leader’s price.
Tools that use it
- Ollama — Ollama. Run open models locally with one command.