Essential cookies keep Instavar working. Optional analytics help us understand how the site is used and link your first-visit source to your Studio account after sign-in, for up to 180 days. Cookie Policy

Manage Cookie Preferences

Service reliability telemetry, including Sentry error monitoring and Vercel Speed Insights, stays enabled so we can secure the product and diagnose failures.

Skip to content
ResearchPlaybooksToolsOpen sourceCase studiesStudio
  1. Home
  2. Blog

Singapore Office - Antiphishing Pte. Ltd.

JTC LaunchPad @ one-north

67 Ayer Rajah Crescent, #02-14

Singapore 139950

© 2026 Instavar. All rights reserved.

Find the angle that travels.

ResearchPlaybooksToolsOpen sourceCase studiesStudioAboutCareers|PrivacyTermsCookiesAI PolicyRights & ConsentReport Abuse

UK Corporate Presence - Litiga Ltd

Registered in England and Wales

Company number 11610573

Registered office: 128 City Road

London, England EC1V 2NX

Incorporated 8 October 2018

Instavar Blog

Make better content decisions.

Practical lessons on video, distribution, and measurement. Start with the question you need to answer, then go deeper when it helps.

Latest notes.

  • MOSS-TTS First Technical Read and Production Reality Check

    A first engineering read of OpenMOSS MOSS-TTS as of February 2026: what is actually released, what benchmark claims are reported, where deployment risk is still high, and how we will follow up after a 24GB feasibility smoke test.

  • NVIDIA NeMo Speech Collection First Technical Read and Production Reality Check

    A first engineering read of NVIDIA NeMo for speech workflows as of February 2026: what is released now, where it fits in an AI video pipeline, where it does not, and what to validate before 24GB production adoption.

GLM-TTS Technical Report for Production Zero-Shot TTS

A practical read of GLM-TTS (arXiv:2512.14291): what is actually novel, what the open benchmark numbers show, what is still internal-only, and how to evaluate fit for production voice workflows.

  • ReStyle-TTS and Relative Style Control in Zero-Shot TTS

    A practical read on ReStyle-TTS (arXiv:2601.03632): what is novel, what the reported results show, what this unlocks for voice workflows, and what still blocks production adoption until code or demo artifacts are published.

  • OCR Benchmark Leaderboard 2026 - Best Models and Workflow Fit

    OCR benchmark leaderboard for 2026 with our 1,651-page OmniDocBench v1.6 results for Jina OCR v1, OvisOCR2, GLM, FireRed, Qwen and other local models on an RTX 3090 Ti, plus reported scores, VRAM and workflow fit.

  • SteadyDancer: Harmonized Human Image Animation with First-Frame Preservation

    SteadyDancer is an open-source human image animation framework that shifts from reference-to-video to image-to-video generation to improve first-frame preservation, temporal coherence, and identity stability under real-world pose and timing misalignment.

  • HunyuanVideo 1.5 - Upgrade Checklist for Production Teams

    A practical checklist for teams evaluating a move from HunyuanVideo v1.0 to HunyuanVideo 1.5, with validation criteria for quality, stability, cost, and rollout safety.

  • CosyVoice 2 vs 3 - Voice Cloning Quality Compared (2026)

    CosyVoice 2 vs CosyVoice 3 voice cloning quality review for 2026. Includes audio samples, zero-shot baseline notes, fine-tuning results, LoRA rerun status, and when to use each model.

  • IMDA NSC Voice Cloning Finetuning Benchmark 2026

    A practical benchmark of CosyVoice, VoxCPM, Qwen3-TTS, and IndexTTS2 fine-tuned on IMDA NSC FEMALE_01, with operational notes for both product teams and engineers.

  • IndexTTS2 Finetuning on IMDA NSC FEMALE_01

    Practical notes from our IndexTTS2 single-speaker finetuning run on IMDA NSC FEMALE_01, including crash recovery, checkpoint retention behavior, and checkpoint selection.

  • Qwen3-TTS LoRA Fine-Tuning - Scale Sweeps, Checkpoints, and Production Defaults

    Qwen3-TTS LoRA fine-tuning guide for custom voices. Covers dataset requirements, 24GB VRAM settings, LoRA scale 0.3 to 0.35, LR override 2e-6, sft_12hz.py bugs, and the companion repo.

  • VoxCPM 1.5 LoRA Finetuning on IMDA NSC FEMALE_01

    Exact run notes for VoxCPM 1.5 LoRA finetuning on IMDA NSC FEMALE_01, including checkpoint selection, prompt/no-prompt behavior, and denoiser tradeoffs.

  • Previous

    Page 5 of 11

    Next