n0p/mlx-video
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • d52e567c56 Enhance precision in denormalization and normalization processes Prince Canuma 2026-01-17 01:14:29 +01:00
  • ecda6d10e5 Merge pull request #9 from Blaizzy/pc/fix-text-encoder Prince Canuma 2026-01-17 01:10:36 +01:00
  • 146f5d2981 Add image-to-video (I2V) conditioning support Prince Canuma 2026-01-17 00:19:52 +01:00
  • 5f86e881d7 Update top_p parameter in sampler function to 1.0 for enhanced sampling control in LTX2TextEncoder Prince Canuma 2026-01-16 21:08:14 +01:00
  • f6e0e5d5a4 Update av_ca_timestep_scale_multiplier to 1000 in model configuration for consistency across modules Prince Canuma 2026-01-16 15:59:22 +01:00
  • e1bff927df Auto-detect timestep_cond from model metadata () Prince Canuma 2026-01-16 14:55:50 +01:00
  • a658911f98 add audio Prince Canuma 2026-01-16 01:15:22 +01:00
  • 81daf3f67d Add prompt enhancement feature to video generation Prince Canuma 2026-01-15 14:31:00 +01:00
  • f5134fa172 adjust gelu and precision Prince Canuma 2026-01-15 12:49:21 +01:00
  • 349a82f763 Refactor GroupNorm3d: Optimize data type handling by casting input, weight, and bias to float32 for consistency and performance Prince Canuma 2026-01-15 04:46:56 +01:00
  • 09c2b460a7 Refactor LTX2VideoDecoder and ResBlockGroup: Change up_blocks and res_blocks from lists to dictionaries for better parameter tracking in MLX Prince Canuma 2026-01-15 03:48:16 +01:00
  • 3fcd8f90be Refactor LTXModel: Change transformer_blocks from list to dictionary Prince Canuma 2026-01-15 03:47:52 +01:00
  • e7067fea11 Refactor LTX2VideoDecoder: Remove redundant comments for residual parameter Prince Canuma 2026-01-14 01:21:43 +01:00
  • 957093c29b use numpy for improved float64 precision and performance Prince Canuma 2026-01-14 00:03:00 +01:00
  • 74af04718d Remove commented-out code and clean up text encoder initialization Prince Canuma 2026-01-13 23:31:54 +01:00
  • ea063f7550 Cast LM weights to bfloat16 Prince Canuma 2026-01-13 23:30:26 +01:00
  • fc6ef20c1b Add custom text encoder with quantization Prince Canuma 2026-01-13 22:56:51 +01:00
  • 01d895bc77 Add frame number validation in video generation and update Gemma3 text encoder to use validated mlx-vlm implementation Prince Canuma 2026-01-13 17:12:11 +01:00
  • 61b003ff2c Revise README for text-to-video generation example Prince Canuma 2026-01-12 17:21:54 +01:00
  • 535dd9a066 Update README for macOS requirements Prince Canuma 2026-01-12 17:19:32 +01:00
  • 070edc0de6 Replace poodles.mp4 with poodles.gif in examples directory Prince Canuma 2026-01-12 17:14:12 +01:00
  • 74f7f22a0f remove ref Prince Canuma 2026-01-12 17:11:05 +01:00
  • 084927c74a Remove optional dependencies section from README Prince Canuma 2026-01-12 16:48:28 +01:00
  • ebe4e74585 Add pre-commit configuration for code formatting and linting with Black, isort, and autoflake Prince Canuma 2026-01-12 16:47:34 +01:00
  • 4f6fc8252c Add example usage to README and enhance console output in generate.py with ANSI colors Prince Canuma 2026-01-12 16:45:09 +01:00
  • 28417fe126 fix git ignore Prince Canuma 2026-01-12 16:35:41 +01:00
  • 54fb1ed076 fix uv lock Prince Canuma 2026-01-12 16:35:15 +01:00
  • c7f94052e8 fix toml Prince Canuma 2026-01-12 16:35:03 +01:00
  • 28d03d8846 setup Prince Canuma 2026-01-12 16:17:45 +01:00
  • 7eac6ae7de Replace imageio with OpenCV for video saving in generate.py; updated default frame count to 100. Prince Canuma 2026-01-12 16:12:41 +01:00
  • 666e1f2e0c Refactor model path handling: moved get_model_path function to utils.py and updated generate.py to use the new import. Prince Canuma 2026-01-12 15:54:32 +01:00
  • 75511a0b17 Remove main.py and refactor video generation logic into generate.py. Prince Canuma 2026-01-12 14:23:02 +01:00
  • 7114b023bd - Refactor video generation script - Introduced argparse for parameter handling, streamlined model loading, and enhanced denoising functions. - Updated VAE weight sanitization for compatibility and improved activation function handling in text projection. - Added support for saving individual frames and refined output video generation process. Prince Canuma 2026-01-12 14:04:53 +01:00
  • d1ca36a315 initial commit (LTX-2) Prince Canuma 2026-01-11 23:48:33 +01:00
  • 9f01d22750 Initial commit Prince Canuma 2025-05-07 12:21:09 +02:00