Written by Jakub Rusinowski · Last updated July 10, 2026
Alibaba's February 2026 flagship generation. Qwen 3.5 immediately topped multiple open-source reasoning benchmarks, scoring 88.4 on GPQA Diamond and 76.4 on SWE-bench Verified. The MoE architecture delivers 8–19× higher decoding throughput vs earlier Qwen3 while supporting 200+ languages. Apache 2.0 licensed for full commercial use.
| Qwen 3.5 7B | Min 5 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3.5:7b |
| Qwen 3.5 14B | Min 10 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3.5:14b |
| Qwen 3.5 32B | Min 20 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3.5:32b |
| Qwen 3.5 122B-A10B (MoE) | Min 14 GB VRAM · Q4_K_M · 128,000 ctx · ollama run qwen3.5:122b-a10b-q4 |
The cheapest GPU that runs Qwen 3.5 (Legacy Listing — Unverified) locally (min 5 GB VRAM) is the Intel Arc B570 (10 GB).
Install Ollama then run: ollama run qwen3.5:7b
Minimum VRAM: 5 GB. For best results use Q4_K_M quantization.
Qwen 3.5 (Legacy Listing — Unverified) needs about 5 GB VRAM at Q4_K_M quantization for its smallest variant. Variants: Qwen 3.5 7B (5 GB, Q4_K_M); Qwen 3.5 14B (10 GB, Q4_K_M); Qwen 3.5 32B (20 GB, Q4_K_M); Qwen 3.5 122B-A10B (MoE) (14 GB, Q4_K_M). On Apple Silicon, unified memory counts toward this requirement.
Yes — Qwen 3.5 (Legacy Listing — Unverified) runs on an RTX 4090 (24 GB) and other 24 GB cards such as the RTX 3090. Smaller variants also fit comfortably on 8–16 GB GPUs at Q4_K_M.
Q4_K_M is the best balance of quality and VRAM for Qwen 3.5 (Legacy Listing — Unverified) in most cases. Choose Q8_0 for near-lossless quality if you have spare VRAM, or smaller quants (Q3/Q2) only when memory is tight.
Install Ollama, then run: ollama run qwen3.5:7b. This downloads Qwen 3.5 (Legacy Listing — Unverified) and starts a local, OpenAI-compatible endpoint — no internet connection is needed after the initial download.