Add Bernini-R-1.3B MLX (single-expert Wan2.1-1.3B renderer)

NOTICE

@@ -0,0 +1,23 @@
+bernini-r-mlx (1.3B tier)
+Apache MLX port of ByteDance Bernini-R-1.3B (the Bernini Renderer, small tier).
+
+This work is licensed under the Apache License, Version 2.0.
+
+It is derived from and depends on the following Apache-2.0 works; their notices
+and attributions are retained here:
+
+- ByteDance/Bernini-R-1.3B — the Bernini Renderer, 1.3B tier (weights + reference
+  inference code).
+  https://github.com/bytedance/Bernini  ·  https://huggingface.co/ByteDance/Bernini-R-1.3B-Diffusers
+  Paper: "Bernini: Latent Semantic Planning for Video Diffusion" (arXiv:2605.22344).
+
+- Wan-AI/Wan2.1-T2V-1.3B — the base DiT, 16-channel causal VAE, and UMT5 text encoder
+  that Bernini-R-1.3B fine-tunes / reuses.  https://github.com/Wan-Video/Wan2.1
+
+- Qwen2.5-VL-7B-Instruct — the Bernini *planner* (NOT used here; not released as weights).
+
+- mlx-video (Blaizzy/mlx-video) — the MLX Wan backbone reused by this port.
+
+Scope note: only the Bernini *Renderer* is open-sourced upstream. The MLLM semantic
+planner (the paper's "latent semantic planning") is not released, so this port runs with
+UMT5 text conditioning only; the planner-feature channel is absent.

README.md

@@ -0,0 +1,81 @@
+---
+license: apache-2.0
+library_name: mlx
+pipeline_tag: text-to-video
+tags:
+  - mlx
+  - text-to-video
+  - video-editing
+  - video-to-video
+  - reference-to-video
+  - wan2.1
+  - bernini
+base_model: ByteDance/Bernini-R-1.3B-Diffusers
+---
+
+# Bernini-R-1.3B (MLX)
+
+Apple MLX port of **[ByteDance/Bernini-R-1.3B](https://huggingface.co/ByteDance/Bernini-R-1.3B-Diffusers)** —
+the **1.3B** tier of ByteDance's Bernini *Renderer*: a Wan2.1-T2V-1.3B-derived video
+generator/editor with **Segment-Aware 3D RoPE** for multi-reference / editing tasks.
+The small tier "performs close to the 14B variant on simple tasks such as style transfer,
+subtitle or watermark removal, and local editing."
+
+Runs on Apple Silicon via [MLX](https://github.com/ml-explore/mlx) + the
+[mlx-video](https://github.com/Blaizzy/mlx-video) Wan backbone.
+
+> **This is the lowest-cost Bernini tier.** For the higher-quality A14B renderer see
+> [`mlx-community/Bernini-R-bf16`](https://huggingface.co/mlx-community/Bernini-R-bf16) /
+> [`-int4`](https://huggingface.co/mlx-community/Bernini-R-int4).
+
+## ⚠️ Scope: renderer only
+
+Only the Renderer ("-R") is open-sourced upstream. The MLLM semantic **planner** (the
+paper's headline "latent semantic planning", a Qwen2.5-VL-7B model) is **not released**.
+This port therefore runs with **UMT5 text conditioning only** — the planner-feature
+channel is absent (and carries no weights in the released checkpoint).
+
+## Architecture
+
+| | |
+|---|---|
+| Backbone | **Wan2.1-T2V-1.3B**, single expert (30L · dim 1536 · 12H · ffn 8960) |
+| Experts | **one** (`skip_transformer_2: true`, `switch_dit_boundary: 0`) |
+| VAE | 16-ch `AutoencoderKLWan` (stock Wan2.1) |
+| Text encoder | UMT5-xxl |
+| Bernini knobs | `shift 3.0`, `use_src_id_rotary_emb` (SA-3D RoPE — **no extra parameters**) |
+
+Differs from the A14B port only by config: a single 1.3B expert instead of the
+high/low-noise A14B pair. There is no expert-boundary switch.
+
+## Tasks
+
+| Task | Description |
+|---|---|
+| `t2v` / `t2i` | text-to-video / image |
+| `r2v` | reference-to-video — generate a subject from up to K reference images (chained APG) |
+| `v2v` | prompt-based video editing (source video injected as conditioning) |
+| `rv2v` | reference + video editing |
+
+## Variants
+
+| Repo | Precision | Transformer | + shared |
+|---|---|---|---|
+| `…-1.3B-bf16` | bfloat16 | 2.6 GB | VAE 0.5 GB · UMT5 10.6 GB |
+| `…-1.3B-int4` | 4-bit (group 64) | 0.8 GB | VAE 0.5 GB · UMT5 10.6 GB |
+
+## Provenance & validation
+
+- Architecture: **stock Wan2.1-T2V-1.3B** (verified — diffusers `WanTransformer3DModel`
+  keys, no extra tensors; matches `mlx-video` `wan21_t2v_1_3b` exactly: 30L/1536/12H/8960).
+  Bernini knobs (`switch_dit_boundary 0`, `shift 3.0`, `use_src_id_rotary_emb`) live in
+  the wrapper config; SA-3D RoPE adds **no parameters**.
+- Converted fp32 → bf16 from `ByteDance/Bernini-R-1.3B-Diffusers`; VAE/UMT5 are the shared
+  stock Wan2.1 components (byte-identical to the A14B port).
+- Validated on the CPU stream: key contract bijective (825 file tensors = model params
+  minus the derived `freqs` rope buffer); **bf16 forward cosine 0.999983 vs source fp32**;
+  int4 per-pass cosine **0.9943** vs bf16 (group 64, ≥0.99 gate).
+
+## License & attribution
+
+Apache-2.0. Derived from ByteDance Bernini-R, Wan2.1 (Wan-AI), and mlx-video. See `NOTICE`.

config.json

@@ -0,0 +1,48 @@
+{
+  "model_type": "t2v",
+  "model_version": "2.1",
+  "patch_size": [
+    1,
+    2,
+    2
+  ],
+  "text_len": 512,
+  "in_dim": 16,
+  "dim": 1536,
+  "ffn_dim": 8960,
+  "freq_dim": 256,
+  "text_dim": 4096,
+  "out_dim": 16,
+  "num_heads": 12,
+  "num_layers": 30,
+  "window_size": [
+    -1,
+    -1
+  ],
+  "qk_norm": true,
+  "cross_attn_norm": true,
+  "eps": 1e-06,
+  "vae_stride": [
+    4,
+    8,
+    8
+  ],
+  "vae_z_dim": 16,
+  "dual_model": false,
+  "boundary": 0.0,
+  "sample_shift": 3.0,
+  "sample_steps": 50,
+  "sample_guide_scale": 5.0,
+  "num_train_timesteps": 1000,
+  "sample_fps": 16,
+  "frame_num": 81,
+  "sample_neg_prompt": "色调艳丽，过曝，静态，细节模糊不清，字幕，风格，作品，画作，画面，静止，整体发灰，最差质量，低质量，JPEG压缩残留，丑陋的，残缺的，多余的手指，画得不好的手部，画得不好的脸部，畸形的，毁容的，形态畸形的肢体，手指融合，静止不动的画面，杂乱的背景，三条腿，背景人很多，倒着走",
+  "max_area": 0,
+  "t5_vocab_size": 256384,
+  "t5_dim": 4096,
+  "t5_dim_attn": 4096,
+  "t5_dim_ffn": 10240,
+  "t5_num_heads": 64,
+  "t5_num_layers": 24,
+  "t5_num_buckets": 32
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:212747be887cfd8b584d0a6f3d83299714af0da411e29716c1820b6d8637956f
+size 2838077537

t5_encoder.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e86ee4199903e00a88dcd43583a43a6eb898cef600e38670f222d7e37d163787
+size 11361845505

vae.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:977530e453dbfabbab31e2972e1577d8d7e2840ba7410c81aa3fd421c0cd7414
+size 507591226