Maestro v1.6.0 Adds MiniMax H3 Full Model Support and Local Video/Audio Generation
cocktailpeanut · x · 2026-08-08
Maestro releases v1.6.0, bringing major updates for MiniMax H3 models.
- Omni Multimodal Generation: Supports generating new video and synchronized audio from ordered image, video, and sound references. References can be reordered and labeled.
- Models & VRAM: Offers 20B Pruned and 33B Full models. Introduces Match Output mode for consumer GPUs and optimizes memory usage.
- Quantization & Turbo LoRA: Adds multiple text encoder quantization options (NVFP4-AWQ, GGUF, etc.) and supports H3 Turbo LoRA.
- Workflow: Omni generation is capped at 14.4 seconds. The First & Last workflow now supports longer videos by continuing from the preceding final frame.
Related event: Jeff Dean Backs AI Startup Sophont as Second Investor(2 posts)→
More from Infra
- Grok claims 816B MoE model runs locally on dual M5 Ultra Mac Studios via MLX — grok · 2026-10-16
- Same 1,000 Cloudflare Durable Objects: $4,050/month cut to $22 with one trick — threepointone · 2026-10-16
- NVIDIA ships FP4-quantized Qwen3.8-Flash for local use, trending on Hugging Face — nvidia · 2026-10-16
- Baidu's Virtual Head Attention tames the KV cache, FLOPs, quality triangle in long-context LLMs — baidu · 2026-10-16
- DuoDiscover agents auto-build fast eval proxies, scoring 2.68x TPU kernel speedup — berkeley_ai · 2026-10-16
- Google Raises 2026 Capital Spending to Up to $205 Billion — emmanuelvivier · 2026-10-16