Ref2V video model roundup: no open model nails both action and character consistency
MastMaithun · reddit · 2026-10-31
A heavy Ref2V user argues current open video models can't do both dynamic action (fights, dance) and reference fidelity. Controlled tests show swapping only the base model moves reference accuracy from 50% to 95-100%, proving prompts aren't the issue. Tested: Hybrid 20-49 (80% reference, meh motion), Hybrid 15-40 (95-100% reference, can't kick), Eros (70%), Singularity (great action, 50% reference). The only usable combo so far is Hybrid 15-40 + Singularity v2 LoRA at 0.4 strength (70% reference). The community still lacks a model with 80%+ reference and solid motion.
More from Multimodal
- Halloween Midjourney SREF share: style code 4068509853 delivers a vintage horror-poster look — tisch_eins · 2026-10-31
- StudioSphere Pulse Launches Free Audio Intelligence MCP Connector for BPM and Key Analysis — modelcontextprotocol · 2026-10-31
- GPT-6 Astra via invideo MCP builds Galaxy Effect end-to-end in a real editor — gen_ericai · 2026-10-31
- Magnific launches Move: conversational agentic motion graphics with every layer editable — charis_ai · 2026-10-31
- Director runs 'vibe directing' with Claude Code: per-shot approvals, $0.82 clips — DavidmComfort · 2026-10-31
- Ostris trains MOSS Image with Primal VAE: 4K at 1080p compute cost — Crazy-Repeat-2006 · 2026-10-31