feat: i2v vision tower + processor + PATCHES (update pipeline)

PATCHES.md

@@ -1,60 +1,153 @@
-# PATCHES.md — MiniMax-H3 MLX 8-bit pipeline modifications
+# PATCHES.md — MiniMax-H3 MLX 8-bit pipeline modifications (complete port)
 
-Two files are modified relative to the upstream `PipeNetwork/minimax-h3-mlx` port
-(base commit `b2f7e4d2b7861cefe68b75e4b59ab81cc4e7c318`, 2026-08-10). Both changes were
-validated with real generations including Turbo LoRA merges (see the validation report in
-`README.md`). Everything else is byte-identical to the base commit.
+All modifications relative to upstream `PipeNetwork/minimax-h3-mlx` base commit `b2f7e4d2b7861cefe68b75e4b59ab81cc4e7c318` (2026-08-10). The first committed patch is `video-lab-8bit-fixes` `7210b93e6df86bf9c7206091c9542b6983c10c30` (2026-08-17, two files); fixes S8/S9/S10 extend the same branch in the working tree (dit.py two-phase, pipeline release + TeaCache hook, text_encoder processor+scatter, teacache.py, generate.py flags, processor/ + vision shards). Everything else is byte-identical to the base commit.
 
 | | |
 |---|---|
 | Base commit | `b2f7e4d2b7861cefe68b75e4b59ab81cc4e7c318` |
-| Branch | `video-lab-8bit-fixes` |
-| Patch commit | `7210b93e6df86bf9c7206091c9542b6983c10c30` |
-| Date | 2026-08-17 |
+| Branch | `video-lab-8bit-fixes` (+ working-tree S8/S9/S10 on `video-lab`) |
+| Patch commit (committed) | `7210b93e6df86bf9c7206091c9542b6983c10c30` |
+| Date (committed) | 2026-08-17 |
+| Working-tree S8/S9/S10 | 2026-08-18–2026-08-19 (pipeline.py, dit.py extension, teacache.py, text_encoder.py B1/B2, generate.py, processor/ + vision shards) |
+| Validated with | Real generations incl. Turbo LoRA merges + I2V encode checks (see README Validation report + tmp/S9_teacache_spike_RESULT.md) |
+| License | Port: Apache-2.0 (PipeNetwork). Weights: MiniMax H3 Community License (`LICENSE`). Turbo LoRA: Apache-2.0 (larryvrh). |
 
-## `minimax_h3_mlx/dit.py` — DiT attention QKV layout fix
+---
 
-- The raw-checkpoint `attn.qkv_proj` rows are **blocked**, not per-head interleaved:
-  `[q | k | v]` in three contiguous parts of `heads * head_dim` columns each. The old code
-  reshaped to `(B, S, heads, 3, head_dim)` and indexed q/k/v on axis -2, which scrambled the
-  heads; the new code splits and reshapes each third independently:
+## (a) `minimax_h3_mlx/dit.py` — DiT attention QKV blocked layout + TeaCache two-phase hooks
 
+**File:** `minimax_h3_mlx/dit.py` (~20 lines → ~230 lines added vs base)
+
+**QKV blocked layout (S7, committed 7210b93, lines ~12–17 docstring + lines 153–165 in `Attention.__call__`):**
+- Raw-checkpoint `attn.qkv_proj` rows are **blocked** `[q | k | v]` — three contiguous thirds of width `heads * head_dim` — not per-head interleaved `[h0: q,k,v][h1: q,k,v]...`. The old code reshaped to `(B, S, heads, 3, head_dim)` and indexed `qkv[...,0]` which scrambled heads. Fix splits and reshapes each third independently (same layout `mlx-serve` consumes via `splitEqual(qkv,3)`; video VAE `to_qkv` remains per-head interleaved — different codebase, different convention):
   ```python
+  # minimax_h3_mlx/dit.py:160-164 (Attention)
   qkv = self.qkv_proj(x)
   q, k, v = mx.split(qkv, 3, axis=-1)
   q = q.reshape(B, S, self.heads, self.head_dim)
   k = k.reshape(B, S, self.heads, self.head_dim)
   v = v.reshape(B, S, self.heads, self.head_dim)
   ```
+- Module docstring (lines 12–17, 157–158) now documents the blocked layout and the VAE contrast.
+
+**TeaCache two-phase forward (S9, working-tree, lines ~335–510):**
+- New helpers `TYPE_CHECKING` import of `ModulationCache`, `_resolve_cache_layer` (lines 347–354), `_prepare_stages` (355–391, shared input projections + token_refiner + packed buffer + temb + adaln indices so the prefix graph is built exactly once), `_run_blocks` (393–409, iterates `blocks[i]` with optional `ModulationCache.get(i)`), `_finish_heads` (411–423, `final_layer.norm_out` + `video_out`/`audio_out` heads).
+- Public two-phase entry points: `forward_to_cache_layer` (425–464, partial forward up to and including `cache_layer` — default last block `len(blocks)-1` — returning `(hidden, ctx)` where `hidden` is pre-final-norm residual after that block and `ctx` carries rotary/temb/adaln/layer bookkeeping) and `finish_from_cache_layer` (466–487, completes from cached `hidden`/`ctx` without recomputing prefix, returns `(video_velocity, audio_velocity)`). `__call__` (489–510) now delegates to `_prepare_stages`/`_run_blocks`/`_finish_heads` and gains `return_hidden` for probing. Saving per skip = suffix `blocks[layer+1..49]` + `final_layer.norm_out` + both heads — with default last block only heads+norm are saved (~<5% wall); moving to block 40 saves ~20% blocks per skip (see S9 spike result: 0/7 skips at last block, no wall saving).
 
-- The module docstring was updated to document the blocked layout — the same layout the
-  mlx-serve runtime consumes via `splitEqual(qkv, 3)` — and to note that the video VAE's
-  `to_qkv` (per-head interleaved) is a different codebase with a different convention.
+---
 
-## `minimax_h3_mlx/text_encoder.py` — 8-bit quantized text encoder loader
+## (b) `minimax_h3_mlx/text_encoder.py` — 8-bit quantized loader + processor fallback (B1) + positional scatter (B2)
 
-- Detects the MLX affine 8-bit pack by the presence of `quant_config.json` in the model
-  directory and quantizes the Qwen3-VL module tree *before* loading weights, so packed
-  U32 weights / scales / biases key up 1:1:
+**File:** `minimax_h3_mlx/text_encoder.py` (+84 lines vs base, ~6 → ~17 imports)
 
+**8-bit quantized loader (S7, committed 7210b93, lines ~64–110 + ~143–180):**
+- Detects MLX affine 8-bit pack by presence of `quant_config.json` (line 71: `self.quantized = (model_dir / "quant_config.json").exists()`). Before loading, quantizes the module tree so packed U32/scales/biases key 1:1 — same recipe `load.py` replays from `quant_config.json`:
   ```python
-  self.quantized = (model_dir / "quant_config.json").exists()
-  ...
+  # lines 90-92
   nn.quantize(self.language, group_size=64, bits=8,
               class_predicate=lambda _path, m: isinstance(m, nn.Linear))
   ```
+- Vision tower: if `load_vision=True` and `vision_quant_config.json` exists, replays its recorded `quantized` paths (89 tensors) with same group/bits (lines 98–110):
+  ```python
+  with open(model_dir / "vision_quant_config.json") as fh: vq = json.load(fh)
+  paths = set(vq.get("quantized", ()))
+  nn.quantize(self.vision, group_size=vq.get("group_size",64), bits=vq.get("bits",8),
+              class_predicate=lambda _p, m: isinstance(m, nn.Linear) and _p in paths)
+  ```
+- Quantized shards loaded raw (no `astype(dtype)`, lines 166–170: `buckets[bucket][path] = tensor` when `self.quantized` else `tensor.astype(dtype)`), preserving packed integers. Only dense path casts.
+- Serve pack omits `model.norm` (never evaluated — H3 conditions pre-norm); loader fabricates zeros like dense converter (lines 176–180):
+  ```python
+  buckets["language"]["norm.weight"] = mx.zeros(tuple(module.norm.weight.shape), dtype=mx.bfloat16)
+  ```
 
-- Quantized shards are loaded raw (no `astype(dtype)`), keeping the packed integers intact;
-  only the dense path casts tensors.
+**Processor property with torch-free fallback (B1, S10, lines 207–233):**
+- `@property processor` (lines 207–233) tries `AutoProcessor.from_pretrained(processor_dir)` (full Qwen3VLProcessor with torch video/image sub-processors). On `ImportError` (no torch/torchvision; `Qwen3VLVideoProcessor` hard-requires them, and transformers 5.15 `AutoImageProcessor` is itself gated), falls back to PIL-only `Qwen2VLImageProcessorPil`:
+  ```python
+  try: self._processor = AutoProcessor.from_pretrained(processor_dir)
+  except Exception:
+      try: image_processor = AutoImageProcessor.from_pretrained(processor_dir)
+      except Exception:
+          from transformers.models.qwen2_vl.image_processing_pil_qwen2_vl import Qwen2VLImageProcessorPil
+          image_processor = Qwen2VLImageProcessorPil.from_pretrained(processor_dir)
+      self._processor = SimpleNamespace(image_processor=image_processor)
+  ```
+  The facade only ever reads `.image_processor` (`build_request` line 251), tokenizer comes from separate `tokenizer` property (lines 197–205), so image-only processor is sufficient for I2V.
 
-- The serve pack omits `model.norm` (never evaluated — H3 conditions pre-norm), so the loader
-  fabricates the same zeros the dense converter wrote when the weight is missing:
+**Encode positional scatter fix (B2, S10, lines 324–344):**
+- `encode()` (lines 306–344) previously did `mx.where(image_mask[...,None], hidden[None], inputs_embeds)` which mis-broadcasts `(1,3096,1)` vs `(1,3072,5120)` because vision hidden rows are compact while image pads sit at arbitrary positions in the longer token row — a **positional scatter**, not a broadcast.
+- Fix validates `hidden.shape[0] == num_image` (lines 330–334), then does positional REPLACE scatter via `scatter_add` (mlx 0.32 has no `nonzero`/`index_put`, only `at[].add`):
+  ```python
+  pos_np = np.nonzero(np.array(image_mask))[0]   # line 339 (mlx has no nonzero, use numpy)
+  pos = mx.array(pos_np)                         # line 340
+  target = inputs_embeds[0]                      # line 341
+  inputs_embeds = target.at[pos].add(hidden.astype(target.dtype) - target[pos])  # line 342
+  inputs_embeds = inputs_embeds[None]            # line 343
+  ```
+  Validated with `tmp/s10_validate_encode.py` (1,3096,5120) finite, 3074 video-tagged rows (start+3072 pads+end). `build_request` (lines 237–277) tags whole vision block as `TAG_VIDEO` (not text) — the DiT AdaLN key.
+
+---
+
+## (c) `minimax_h3_mlx/pipeline.py` — `release_text_encoder` headroom (S8/S8b) + TeaCache hook (S9) + checkpoint resume
 
+**File:** `minimax_h3_mlx/pipeline.py` (+~205 lines vs base)
+
+**Release text encoder (S8/S8b, lines 59–73 `__init__`, 77–118 `from_pretrained`, 228–331 `__call__`, 470–532 `_release_text_encoder_now`):**
+- `MiniMaxH3Pipeline.__init__` gains `release_text_encoder: bool` (line 66, stored as `self._release_text_encoder` line 73). `from_pretrained` threads it (lines 84, 118).
+- In `__call__` (lines 328–331), right after `prompt_embeds` built (line 1, text conditioning + vision rows already encoded, step-4 noise already drawn — trajectory fixed), opt-in `release_text_encoder or self._release_text_encoder` drops the encoder:
   ```python
-  buckets["language"]["norm.weight"] = mx.zeros(
-      tuple(module.norm.weight.shape), dtype=mx.bfloat16
-  )
+  if (release_text_encoder or self._release_text_encoder) and getattr(self, "text_encoder", None) is not None:
+      self._release_text_encoder_now()
   ```
+- `_release_text_encoder_now` (lines 470–532) drops `self.text_encoder` (`del` + `gc.collect()` pattern), sizes correctly via `mlx.utils.tree_flatten` (lines 484,496,510 — previous `module.parameters().values()` summed dicts not arrays → 0.0 GB), calls `mx.clear_cache()` to return unified Metal memory (line 528), logs `released text encoder after conditioning (freeing ~N GB)` (line 532). Second call is no-op (line 471 guard). Safe: inner estimator errors just leave freed-size as "(sizing indeterminado)" while delete still happens. Measured: frees ~27.5 GB / ~22 GB resident before denoise loop; pipeline then only touches DiT+VAEs.
+- CLI: `scripts/generate.py` `--release-encoder` (line 56) → `pipeline(release_text_encoder=args.release_encoder)` (line 96).
+- Checkpoint resume (bonus, lines 371–389): `resume_from`/`checkpoint_dir` floats inside the loop (not a pipeline-level patch for HF, but present in `pipeline.py` working tree).
+
+**TeaCache hook (S9, lines 43, 228–230, 342–410):**
+- Imports `TeaCacheConfig/TeaCacheController` (line 43), `__call__` gains `teacache`, `teacache_config`, `teacache_layer` (lines 228–230 docstring 244–252). Controller instantiated when `teacache=True` (lines 342–350, resolves `tc_layer = num_layers-1 if None else int`, validates range). Per-step two-phase: `hidden, ctx = dit.forward_to_cache_layer(..., cache_layer=tc_layer)` → `mx.eval(hidden)` → `skip = controller.decide(i, hidden)` → on skip reuse `cached_preds = (video_pred, audio_pred)` joint (lines 391–399), else `finish_from_cache_layer` and cache preds (lines 406–410). Log: `step N/T teacache skip (rel_l1=..., thresh=...)` (lines 398–401). Default last-block hook measured **0/7 skips** at 768x448 turbo8 (see teacache.py below → OFF recommended).
+
+**Other:** `load.py` one-line import fix for quantized path (line 1 added), `scripts/generate.py` adds `--teacache*`/`--release-encoder` flags (lines 45–56, 85–96).
+
+---
+
+## (d) `minimax_h3_mlx/teacache.py` — two-phase TeaCache controller (S9, new file)
+
+**File:** `minimax_h3_mlx/teacache.py` (11 KB, new — 223 lines)
+
+- **Intent:** feature-caching (ByteDance/ali-vilab line) — each step runs partial forward to a probe hidden, compares against previous step's feature, reuses previous `(video, audio)` velocity if similar. Joint reuse (both modalities from one forward must be reused together). Noise/keyframe sampling untouched → seed still fixes trajectory.
+- **`TeaCacheConfig` (lines 40–71):** `rel_l1_thresh=0.2` (default, ref 0.15 for 10s 544x960), `metric='rel_l1'|'cosine'`, `start_at_step=3`, `compute_last_step=True` (protects first 3 and last step), `cache_type='avg'|'last'`. `__post_init__` validates.
+- **Pure helpers (lines 87–134):** `relative_l1_distance` (mean|curr-prev|/mean|prev|, lines 87–100), `cosine_similarity` (lines 103–113), `teacache_gate` (1-min(dist,1) for rel_l1 so higher=more similar, lines 115–124), `teacache_should_skip` (lines 126–134: `rel_l1 <= thresh` or `cosine >= thresh`).
+- **`TeaCacheController` (lines 148–242):** stateful per-run (`total_steps`, `cached_feature`, `computed/skipped`, `by_step`). `decide(step_index, feature)` (lines 171–210): `must_compute` if `step==0` or `step<start` or `cache is None` or `last step` → prime cache; else compute `metric_value` (cosine or rel_l1 distance), `skip = metric >=/<= thresh`, smoothing `avg` folds current into cache as `(prev+curr)/2` on skip, else replaces. Stats: `last_gate` (similarity-domain), `last_metric` (raw distance/similarity for honest log), `metric_name`, `to_stats()`.
+- **Pipeline integration:** see (c) above; `dit.py` hooks provide the probe feature. **Measurement (M4 Max, 768x448 turbo8, 7 forwards):** default last-block `thresh 0.2` → 0 skips (+9.7% overhead), `0.35` → 1/7 skip (rel_l1 0.297) still no wall saving (partial forward already 49/50 blocks, only `final_layer.norm_out`+heads saved). `tmp/S9_teacache_spike_RESULT.md` — **verdict: TeaCache OFF by default**; aggressive needs earlier hook `layer 40` + `thresh 0.25–0.35` for double-digit % (trades quality).
+
+---
+
+## (e) `processor/` + vision shards (S10 I2V)
+
+**Files added to checkpoint (not code patches but data required for I2V):**
+- `processor/` (7 files, ~11.6 MB): `preprocessor_config.json` (390 B), `video_preprocessor_config.json` (385 B), `chat_template.json` (5.5 KB), `tokenizer_config.json` (11 KB), `tokenizer.json` (6.7 MB), `vocab.json` (2.6 MB), `merges.txt` (1.6 MB) — copied from `models/minimax-h3-ckpt/processor/` (`Qwen3VLProcessor`, `Qwen2VLImageProcessorFast` + `Qwen3VLVideoProcessor` configs; runtime uses PIL fallback per (b) B1).
+- `text_encoder/model-00005-of-00008-vision.safetensors` (291,980,081 B), `model-00006-of-00008-vision.safetensors` (341,371,441 B), `model-00007-of-00008-vision.safetensors` (129,267,789 B) — 529 `model.visual.*` tensors (vision tower: depth 27, hidden 1152, patch 16, merge 2, out 5120). Built by `scripts/rebuild_h3_text_encoder_vision_8bit.py` which transposes `patch_embed.proj.weight` from torch `(O,C,kD,kH,kW)` to MLX channels-last and quantizes 89 paths.
+- `text_encoder/vision_quant_config.json` (2,461 B, 89 quantized paths: `blocks.*.attn.qkv/proj`, `blocks.*.mlp.linear_fc1`, `merger.*`, `deepstack_merger_list.*`).
+
+**Why this matters:** without `processor/` the pipeline cannot build `pixel_values`/`image_grid_thw` (I2V falls back to T2V). Without vision shards `text_encoder.encode(prompt, images)` raises `load_vision=False` or missing-tensor error. With both, `encode()` (b-B2) scatters 3072 vision patches into `inputs_embeds` and hits the DiT as `TAG_VIDEO` rows, then `pipeline._encode_keyframes` patches conditioning latents — FL2VA keyframe path (anchors `first`/`last`) validated end-to-end (see README `## Image-to-Video`).
+
+---
+
+## Other minor port fixes
+
+- `minimax_h3_mlx/load.py`: quantization structure replay for DiT (already in committed 7210b93) — one-line import guard retained.
+- `scripts/generate.py`: flags `--teacache`, `--teacache-thresh` (default 0.2), `--teacache-metric` (rel_l1/cosine), `--teacache-layer`, `--teacache-start` (default 3), `--release-encoder` (all S8/S9).
+- `tests/test_release_text_encoder.py`, `tests/test_teacache.py`: unit/mocked tests for (c)/(d) (not shipped in HF package, kept in repo).
+
+## Reproducing the diff
+
+```bash
+git -C models/minimax-h3-mlx diff b2f7e4d2 -- minimax_h3_mlx/dit.py minimax_h3_mlx/text_encoder.py minimax_h3_mlx/pipeline.py minimax_h3_mlx/teacache.py scripts/generate.py
+# committed two-file patch:
+git -C models/minimax-h3-mlx show 7210b93 --stat
+```
+
+## License
 
-- Full patch: `/tmp/h3-port-patches.patch` in the source workspace (`git diff` of the two
-  files, 34 insertions / 6 deletions).
+- **Weights**: MiniMax H3 Community License (included as `LICENSE`) — not open source; territorial exclusions apply (EU/UK/KR/US excluded per mlx-serve pack note).
+- **Port code**: Apache-2.0 (PipeNetwork/minimax-h3-mlx).
+- **Turbo LoRA**: Apache-2.0 (larryvrh/MiniMax-H3-Turbo-Lora).

README.md

@@ -57,6 +57,45 @@ The pipeline needs only this repository: `-c` supplies the VAEs, text encoder an
 `-t` points at the quantized DiT. Native canvas is a 768px short edge (e.g. 1344x768 16:9);
 smaller canvases are off-distribution and degrade quickly.
 
+## Image-to-Video (FL2VA)
+
+This repository is **I2V-capable** (partition `fl2va`, tasks `t2va` + `fl2va` in `model_index.json`). The image is **optional** — present as a keyframe conditioning row via `--image` / `--anchor`, or omitted for pure T2V.
+
+**Processor + vision tower required for I2V.** The `processor/` directory (7 files) and the 3 vision shards (`text_encoder/model-00005..00007-of-00008-vision.safetensors` + `vision_quant_config.json`, 529 `model.visual.*` tensors, 89 quantized) must be present. Without them the pipeline cannot encode images (falls back to T2V only).
+
+**Canonical I2V command (validated):**
+
+```bash
+# 704×544 portrait (height 704 width 544) or 512×384 lightweight variant — both portrait 9:16 / 4:3
+# Bridge image: inputs/bridge-20260813-181546-krea2_turbo-selfie-style-vertical-shot-a-young-woman-in-h.png
+PROMPT=$(cat <<'PROMPT_EOF'
+Vertical 9:16 TikTok-style UGC selfie video, handheld smartphone feel, natural indoor daylight near a window. A friendly creator speaks directly to camera with natural blinking, subtle head nods, and a warm smile. Add small human imperfections: a tiny hesitation, a soft breath, a quick smile mid-sentence, and a micro-pause before the last line. Realistic skin texture, stable identity, no face warping, minimal flicker, clean audio with natural room tone.
+
+No subtitles. No on-screen text. No logos. No watermarks.
+
+The creator says (exactly, with the same pacing and hesitations):
+"Okay, entonces… eh… un datazo. Si estás trabado con tu código, solo da el primer pasito… como, abre tu terminal y escribe fran. (sonríe) Así de simple. Te vas a sorprender de lo rápido que todo se vuelve más fácil."
+PROMPT_EOF
+)
+
+.venv/bin/python scripts/generate.py "$PROMPT" \
+  -c <this repo> -t <this repo>/transformer -s 8 --seed 1996783985 \
+  --height 704 --width 544 -d 5 \
+  --image inputs/bridge-20260813-181546-krea2_turbo-selfie-style-vertical-shot-a-young-woman-in-h.png \
+  --anchor first --release-encoder \
+  -o i2v-704x544.mp4
+
+# Lightweight alternative (faster, same identity):
+.venv/bin/python scripts/generate.py "$PROMPT" \
+  -c <this repo> -t <this repo>/transformer -s 8 --seed 1996783985 \
+  --height 512 --width 384 -d 5 \
+  --image inputs/bridge-20260813-181546-krea2_turbo-selfie-style-vertical-shot-a-young-woman-in-h.png \
+  --anchor first --release-encoder \
+  -o i2v-384x512.mp4
+```
+
+Valid `anchors` are `first` and `last` (FL2VA). Omit `--image`/`--anchor` for text-to-video. The image is consumed by `text_encoder` via `processor/` (`Qwen2VLImageProcessorPil` fallback when torch absent) and the vision tower, then patched as video-conditioning rows (see `minimal I2V encode` in `PATCHES.md`).
+
 ## Example outputs
 
 ![MiniMax-H3 MLX comparison — 8-bit vs turbo variants](images/comparative.png)
@@ -77,7 +116,9 @@ Turbo variants: see rows 2-3 above (a 1344x768 4-bit video re-render is planned)
 | Path | Role | Representation |
 |---|---|---|
 | `transformer/transformer.safetensors` | 33B DiT (joint video+audio) | 8-bit affine, group 64 (AdaLN 8-bit), ~35.3 GB |
-| `text_encoder/` (5 shards) | Qwen3-VL-32B conditioner (64 layers, truncated to layer 50) | 8-bit quantized, ~27.5 GB |
+| `text_encoder/` (5+3 shards) | Qwen3-VL-32B conditioner (64 layers, truncated to layer 50) | 8-bit quantized, ~27.5 GB language (5 shards) + ~0.76 GB vision (3 shards `model-00005..00007-of-00008-vision.safetensors`, 529 `model.visual.*` tensors) |
+| `text_encoder/vision_quant_config.json` | Vision tower quantization recipe (affine 8-bit g64) | json, 89 quantized tensors (`blocks.*`, `merger.*`, `deepstack_merger_list.*`) |
+| `processor/` | Qwen3-VL processor (image preprocessing for I2V) | 7 files: `preprocessor_config.json`, `video_preprocessor_config.json`, `chat_template.json`, `tokenizer.json`, `tokenizer_config.json`, `vocab.json`, `merges.txt` (~11.6 MB) |
 | `video_vae/` | Tiled causal video VAE (17-frame chunks, latents_mean/std) | fp16/bf16, 5.2 GB |
 | `audio_vae/` | DAC encoder + BigVGAN vocoder, stereo 32 kHz | fp32, 0.6 GB |
 | `tokenizer/` | Qwen3-VL tokenizer | json/vocab/merges |
@@ -88,10 +129,12 @@ Turbo variants: see rows 2-3 above (a 1344x768 4-bit video re-render is planned)
 ## Memory and speed (measured, M4 Max 68.7 GB)
 
 - Resident during generation: DiT ~21.5 GB + text encoder ~22 GB + VAEs ~6 GB (AdaLN projections
-  are precomputed and dropped, freeing ~13.8 GB).
-- 768x448, 16 steps: **33.9 min** (119 s/step), peak RSS 24.7 GB, no swap.
+  are precomputed and dropped, freeing ~13.8 GB). With `--release-encoder`, the text encoder is **freed after conditioning** (`prompt_embeds` already built, before the denoise loop), freeing ~27.5 GB (log: `released text encoder after conditioning (freeing ~27.5 GB)` + `mx.clear_cache()` + `tree_flatten` sizing). Peak RSS drops by ~22 GB resident; overhead <1 s.
+- 768x448, 16 steps: **33.9 min** (119 s/step), peak RSS 24.7 GB, no swap. With turbo merge: ~24 min (8 steps). With **turbo8** (pure 8-bit, no merge) at 768x448, 8 steps (7 forwards): **~18.5 min (139 s/step)** vs turbo merge **251 s/step** — **~1.6–1.8× per-pixel** speedup (same canvas, fewer denoise forwards).
 - 1344x768 (native 16:9), 8 steps: **~2 h** (996 s/step), peak ~57 GB + compressed memory; requires
   an otherwise idle machine (jetsam kills it under heavy ambient load).
+- 704×544 I2V (portrait 9:16) turbo merge, 8 steps (7 forwards): **~33 min (251 s/step)**; turbo8: **~139 s/step** (measured 768x448 baseline, scales linearly per pixel). Lightweight 384×512 I2V is proportionally faster (off-distribution, useful for wiring checks).
+- TeaCache: measured **0 skips / 7 forwards** with default last-block hook (`--teacache`, `thresh 0.2`, `start 3`, `compute_last_step`) at 768x448 turbo8 — probe feature pre-final-norm of last block (block 49/50) barely moves, so gate `rel_l1 <= 0.2` never fires; with `thresh 0.35` only 1/7 skips and still **no wall saving** (partial forward already traverses 49/50 blocks, only saves final norm + heads). Overhead +8–10% without skip. **Recommendation: keep TeaCache OFF by default**; if wall saving is needed, use an earlier hook `--teacache-layer 40` with `--teacache-thresh 0.25–0.35` (trades quality for ~20% block saving per skip). See `PATCHES.md` and `tmp/S9_teacache_spike_RESULT.md`.
 - Attention is dense (MiniMax has not released sparse attention); quantization does not reduce the
   attention cost — it exists to fit memory.
 
@@ -105,6 +148,8 @@ was judged by an independent vision-capable model on extracted frames (not stati
 | 8-bit (this repo) | 768x448, 16 steps | 33.9 min | **9/10** — coherent fox walking a mossy log; stable background; no melting |
 | 8-bit + Turbo LoRA | 768x448, 8 steps | ~24 min | High confidence — "impressive for a 4-step turbo LoRA" |
 | 4-bit + Turbo LoRA | 1344x768, 8 steps | 2 h 3 min | **95%** — full leap arc; no melting, no banding (frames; mp4 mux was a local script bug, not the model) |
+| I2V E2E canonical (FL2VA, turbo merge) | 544×704 (704h×544w) 9:16, 8 steps (7 forwards), 5 s, seed 1996783985, `--image bridge-...png --anchor first --release-encoder` | ~33 min (251 s/step) | Keyframe anchored, identity preserved; requires `processor/` + vision tower (529 tensors); image optional (FL2VA otherwise uses key) |
+| I2V Turbo8 (pure 8-bit, FL2VA) | 768×448, 8 steps (7 forwards), 5 s | ~18.5 min (139 s/step, ~1.6–1.8× per-pixel vs turbo merge) | Same I2V prompt/image, pure 8-bit DiT without LoRA merge; faster, same wiring |
 
 Key findings:
 - The VAE decoder is **not** the source of artifacts: parity with the reference diffusers
@@ -119,7 +164,7 @@ Key findings:
 
 ## Turbo LoRA
 
-`turbo_lora_4step_ema.safetensors` is the 4-step EMA distilled LoRA from
+`turbo_lora_4step_ema.safetensors` (also present as `turbo_lora.safetensors` — hardlink alias, same 779,849,816 bytes) is the 4-step EMA distilled LoRA from
 [larryvrh/MiniMax-H3-Turbo-Lora](https://huggingface.co/larryvrh/MiniMax-H3-Turbo-Lora)
 (Apache-2.0; sha256 `5a6eeba1…`). Merge it into a transformer copy:
 
@@ -132,9 +177,7 @@ Key findings:
 
 ## Patches vs upstream
 
-`PATCHES.md` documents the two modified files (branch `video-lab-8bit-fixes`, commit `7210b93e6df86bf9c7206091c9542b6983c10c30`):
-an 8-bit quantized text-encoder loader path and DiT fixes validated with the turbo merges. The
-remaining pipeline is byte-identical to the base commit.
+`PATCHES.md` documents **five** patch families on branch `video-lab` over base commit `b2f7e4d2` (2026-08-10): (a) DiT QKV blocked layout, (b) text-encoder 8-bit + processor fallback + positional scatter fix, (c) `release_text_encoder` headroom, (d) two-phase TeaCache, (e) I2V processor + vision shards. See `PATCHES.md` for per-file details and line references. The remaining pipeline is otherwise byte-identical to the base commit.
 
 ## License
 

processor/chat_template.json

@@ -0,0 +1,3 @@
+{
+    "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {%- if messages[0].content is string %}\n            {{- messages[0].content }}\n        {%- else %}\n            {%- for content in messages[0].content %}\n                {%- if 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' }}\n        {%- if messages[0].content is string %}\n            {{- messages[0].content }}\n        {%- else %}\n            {%- for content in messages[0].content %}\n                {%- if 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- for message in messages %}\n    {%- if message.role == \"user\" %}\n        {{- '<|im_start|>' + message.role + '\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content in message.content %}\n                {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                    {%- set image_count.value = image_count.value + 1 %}\n                    {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n                    <|vision_start|><|image_pad|><|vision_end|>\n                {%- elif content.type == 'video' or 'video' in content %}\n                    {%- set video_count.value = video_count.value + 1 %}\n                    {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n                    <|vision_start|><|video_pad|><|vision_end|>\n                {%- elif 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role + '\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content_item in message.content %}\n                {%- if 'text' in content_item %}\n                    {{- content_item.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and message.content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content in message.content %}\n                {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                    {%- set image_count.value = image_count.value + 1 %}\n                    {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n                    <|vision_start|><|image_pad|><|vision_end|>\n                {%- elif content.type == 'video' or 'video' in content %}\n                    {%- set video_count.value = video_count.value + 1 %}\n                    {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n                    <|vision_start|><|video_pad|><|vision_end|>\n                {%- elif 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n"
+}

processor/preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 16777216,
+        "shortest_edge": 65536
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "image_processor_type": "Qwen2VLImageProcessorFast"
+}

processor/tokenizer_config.json

@@ -0,0 +1,246 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151665": {
+      "content": "<tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151666": {
+      "content": "</tool_response>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151667": {
+      "content": "<think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151668": {
+      "content": "</think>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>",
+    "<d>",
+    "</d>",
+    "<|cutoff|>",
+    "<|lyrics_start|>",
+    "<|lyrics_end|>",
+    "<|caption_start|>",
+    "<|caption_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{%- if tools %}\n    {{- '<|im_start|>system\\n' }}\n    {%- if messages[0].role == 'system' %}\n        {%- if messages[0].content is string %}\n            {{- messages[0].content }}\n        {%- else %}\n            {%- for content in messages[0].content %}\n                {%- if 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '\\n\\n' }}\n    {%- endif %}\n    {{- \"# Tools\\n\\nYou may call one or more functions to assist with the user query.\\n\\nYou are provided with function signatures within <tools></tools> XML tags:\\n<tools>\" }}\n    {%- for tool in tools %}\n        {{- \"\\n\" }}\n        {{- tool | tojson }}\n    {%- endfor %}\n    {{- \"\\n</tools>\\n\\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\\n<tool_call>\\n{\\\"name\\\": <function-name>, \\\"arguments\\\": <args-json-object>}\\n</tool_call><|im_end|>\\n\" }}\n{%- else %}\n    {%- if messages[0].role == 'system' %}\n        {{- '<|im_start|>system\\n' }}\n        {%- if messages[0].content is string %}\n            {{- messages[0].content }}\n        {%- else %}\n            {%- for content in messages[0].content %}\n                {%- if 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- endif %}\n{%- endif %}\n{%- set image_count = namespace(value=0) %}\n{%- set video_count = namespace(value=0) %}\n{%- for message in messages %}\n    {%- if message.role == \"user\" %}\n        {{- '<|im_start|>' + message.role + '\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content in message.content %}\n                {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                    {%- set image_count.value = image_count.value + 1 %}\n                    {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n                    <|vision_start|><|image_pad|><|vision_end|>\n                {%- elif content.type == 'video' or 'video' in content %}\n                    {%- set video_count.value = video_count.value + 1 %}\n                    {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n                    <|vision_start|><|video_pad|><|vision_end|>\n                {%- elif 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"assistant\" %}\n        {{- '<|im_start|>' + message.role + '\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content_item in message.content %}\n                {%- if 'text' in content_item %}\n                    {{- content_item.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {%- if message.tool_calls %}\n            {%- for tool_call in message.tool_calls %}\n                {%- if (loop.first and message.content) or (not loop.first) %}\n                    {{- '\\n' }}\n                {%- endif %}\n                {%- if tool_call.function %}\n                    {%- set tool_call = tool_call.function %}\n                {%- endif %}\n                {{- '<tool_call>\\n{\"name\": \"' }}\n                {{- tool_call.name }}\n                {{- '\", \"arguments\": ' }}\n                {%- if tool_call.arguments is string %}\n                    {{- tool_call.arguments }}\n                {%- else %}\n                    {{- tool_call.arguments | tojson }}\n                {%- endif %}\n                {{- '}\\n</tool_call>' }}\n            {%- endfor %}\n        {%- endif %}\n        {{- '<|im_end|>\\n' }}\n    {%- elif message.role == \"tool\" %}\n        {%- if loop.first or (messages[loop.index0 - 1].role != \"tool\") %}\n            {{- '<|im_start|>user' }}\n        {%- endif %}\n        {{- '\\n<tool_response>\\n' }}\n        {%- if message.content is string %}\n            {{- message.content }}\n        {%- else %}\n            {%- for content in message.content %}\n                {%- if content.type == 'image' or 'image' in content or 'image_url' in content %}\n                    {%- set image_count.value = image_count.value + 1 %}\n                    {%- if add_vision_id %}Picture {{ image_count.value }}: {% endif -%}\n                    <|vision_start|><|image_pad|><|vision_end|>\n                {%- elif content.type == 'video' or 'video' in content %}\n                    {%- set video_count.value = video_count.value + 1 %}\n                    {%- if add_vision_id %}Video {{ video_count.value }}: {% endif -%}\n                    <|vision_start|><|video_pad|><|vision_end|>\n                {%- elif 'text' in content %}\n                    {{- content.text }}\n                {%- endif %}\n            {%- endfor %}\n        {%- endif %}\n        {{- '\\n</tool_response>' }}\n        {%- if loop.last or (messages[loop.index0 + 1].role != \"tool\") %}\n            {{- '<|im_end|>\\n' }}\n        {%- endif %}\n    {%- endif %}\n{%- endfor %}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant\\n' }}\n{%- endif %}\n",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 262144,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

processor/video_preprocessor_config.json

@@ -0,0 +1,21 @@
+{
+    "size": {
+        "longest_edge": 25165824,
+        "shortest_edge": 4096
+    },
+    "patch_size": 16,
+    "temporal_patch_size": 2,
+    "merge_size": 2,
+    "image_mean": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "image_std": [
+        0.5,
+        0.5,
+        0.5
+    ],
+    "processor_class": "Qwen3VLProcessor",
+    "video_processor_type": "Qwen3VLVideoProcessor"
+}

text_encoder/model-00005-of-00008-vision.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3e4592712ef68cba15eb83fd730fbf2176ebe8c991b3240be457913c03be39a5
+size 291980081

text_encoder/model-00006-of-00008-vision.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d943fb9b1b23a476bba0ccb4b6b3fac4c8c5174427dc0e9dffbffc777be530d
+size 341371441

text_encoder/model-00007-of-00008-vision.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ec8e63c80cc27069e9ac3699dd2c81690cf67f6322d669b9fe48f21b6c1b4ba2
+size 129267789

text_encoder/vision_quant_config.json

@@ -0,0 +1,96 @@
+{
+ "bits": 8,
+ "group_size": 64,
+ "mode": "affine",
+ "quantized": [
+  "blocks.0.attn.proj",
+  "blocks.0.attn.qkv",
+  "blocks.0.mlp.linear_fc1",
+  "blocks.1.attn.proj",
+  "blocks.1.attn.qkv",
+  "blocks.1.mlp.linear_fc1",
+  "blocks.10.attn.proj",
+  "blocks.10.attn.qkv",
+  "blocks.10.mlp.linear_fc1",
+  "blocks.11.attn.proj",
+  "blocks.11.attn.qkv",
+  "blocks.11.mlp.linear_fc1",
+  "blocks.12.attn.proj",
+  "blocks.12.attn.qkv",
+  "blocks.12.mlp.linear_fc1",
+  "blocks.13.attn.proj",
+  "blocks.13.attn.qkv",
+  "blocks.13.mlp.linear_fc1",
+  "blocks.14.attn.proj",
+  "blocks.14.attn.qkv",
+  "blocks.14.mlp.linear_fc1",
+  "blocks.15.attn.proj",
+  "blocks.15.attn.qkv",
+  "blocks.15.mlp.linear_fc1",
+  "blocks.16.attn.proj",
+  "blocks.16.attn.qkv",
+  "blocks.16.mlp.linear_fc1",
+  "blocks.17.attn.proj",
+  "blocks.17.attn.qkv",
+  "blocks.17.mlp.linear_fc1",
+  "blocks.18.attn.proj",
+  "blocks.18.attn.qkv",
+  "blocks.18.mlp.linear_fc1",
+  "blocks.19.attn.proj",
+  "blocks.19.attn.qkv",
+  "blocks.19.mlp.linear_fc1",
+  "blocks.2.attn.proj",
+  "blocks.2.attn.qkv",
+  "blocks.2.mlp.linear_fc1",
+  "blocks.20.attn.proj",
+  "blocks.20.attn.qkv",
+  "blocks.20.mlp.linear_fc1",
+  "blocks.21.attn.proj",
+  "blocks.21.attn.qkv",
+  "blocks.21.mlp.linear_fc1",
+  "blocks.22.attn.proj",
+  "blocks.22.attn.qkv",
+  "blocks.22.mlp.linear_fc1",
+  "blocks.23.attn.proj",
+  "blocks.23.attn.qkv",
+  "blocks.23.mlp.linear_fc1",
+  "blocks.24.attn.proj",
+  "blocks.24.attn.qkv",
+  "blocks.24.mlp.linear_fc1",
+  "blocks.25.attn.proj",
+  "blocks.25.attn.qkv",
+  "blocks.25.mlp.linear_fc1",
+  "blocks.26.attn.proj",
+  "blocks.26.attn.qkv",
+  "blocks.26.mlp.linear_fc1",
+  "blocks.3.attn.proj",
+  "blocks.3.attn.qkv",
+  "blocks.3.mlp.linear_fc1",
+  "blocks.4.attn.proj",
+  "blocks.4.attn.qkv",
+  "blocks.4.mlp.linear_fc1",
+  "blocks.5.attn.proj",
+  "blocks.5.attn.qkv",
+  "blocks.5.mlp.linear_fc1",
+  "blocks.6.attn.proj",
+  "blocks.6.attn.qkv",
+  "blocks.6.mlp.linear_fc1",
+  "blocks.7.attn.proj",
+  "blocks.7.attn.qkv",
+  "blocks.7.mlp.linear_fc1",
+  "blocks.8.attn.proj",
+  "blocks.8.attn.qkv",
+  "blocks.8.mlp.linear_fc1",
+  "blocks.9.attn.proj",
+  "blocks.9.attn.qkv",
+  "blocks.9.mlp.linear_fc1",
+  "deepstack_merger_list.0.linear_fc1",
+  "deepstack_merger_list.0.linear_fc2",
+  "deepstack_merger_list.1.linear_fc1",
+  "deepstack_merger_list.1.linear_fc2",
+  "deepstack_merger_list.2.linear_fc1",
+  "deepstack_merger_list.2.linear_fc2",
+  "merger.linear_fc1",
+  "merger.linear_fc2"
+ ]
+}

turbo_lora.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5a6eeba171cf183020a4ad48774bb2968f29f8168afd6ec17a04987f3528b4ea
+size 779849816