AI article
Kandinsky 6.0 on a 16 GB GPU: the official recipe runs out of memory, and only the 29B model passed my Spanish test
Community description: Kandinsky 6.0 generates five seconds of video with its audio, voices included, from a text prompt...
Dev.to | Oct 14, 2026 | Efrain Garay
Automated excerpt
On a rented 16 GB RTX 4080, the README recipe (per-component CPU offload) ran out of memory with the distilled Lite model; the Qwen2. 5-VL text encoder files alone add up to 15. 45 GiB. An 8-bit text encoder and sequential offload failed too. PyTorch allocated at most 11. 24 GiB; a vertical clip took about 141 s; the process reached 45. 9 GiB of system RAM.
Selected automatically from source text; not independently written or fact-checked. Read the original for full context.