The dataset viewer is not available for this split.
Error code: StreamingRowsError
Exception: CastError
Message: Couldn't cast
schema_version: string
version: string
format: string
format_authorized_by: string
paid_generation_authorized: bool
publication_authorized: bool
models: list<item: string>
child 0, item: string
volume: struct<id: string, region: string, models_root: string>
child 0, id: string
child 1, region: string
child 2, models_root: string
booths: list<item: struct<id: string, name: string, skill: string, setup: list<item: string>, action: list<i (... 1145 chars omitted)
child 0, item: struct<id: string, name: string, skill: string, setup: list<item: string>, action: list<item: string (... 1133 chars omitted)
child 0, id: string
child 1, name: string
child 2, skill: string
child 3, setup: list<item: string>
child 0, item: string
child 4, action: list<item: string>
child 0, item: string
child 5, ending: list<item: string>
child 0, item: string
child 6, continuity: list<item: string>
child 0, item: string
child 7, order: int64
child 8, seeds: list<item: int64>
child 0, item: int64
child 9, canonical_brief: string
child 10, prompt_variants: struct<ltx-2.5-distilled-bf16: struct<prompt: string, method: string, sources: list<item: string>, s (... 855 chars omitted)
child 0, ltx-2.5-distilled-bf16: struct<prompt: string, method: string, sources: list<item: string>, semantic_review: struct<type: st (... 191 chars omitted)
child 0, prompt: string
...
child 0, width: int64
child 1, height: int64
child 2, frames: int64
child 3, fps: string
child 2, wan2.2-t2v-a14b: struct<width: int64, height: int64, frames: int64, fps: string>
child 0, width: int64
child 1, height: int64
child 2, frames: int64
child 3, fps: string
child 3, image_digest: string
child 4, software_commits: struct<ltx-2.5-distilled-bf16: string, kandinsky-6-lite-distill-5s: string, wan2.2-t2v-a14b: string>
child 0, ltx-2.5-distilled-bf16: string
child 1, kandinsky-6-lite-distill-5s: string
child 2, wan2.2-t2v-a14b: string
child 5, endpoint_ids: null
child 6, manifest: string
child 7, manifest_sha256: string
spend_envelope: struct<authorized: bool, incremental_testing_ceiling_usd: int64, preflight_ceiling_usd: int64, conte (... 279 chars omitted)
child 0, authorized: bool
child 1, incremental_testing_ceiling_usd: int64
child 2, preflight_ceiling_usd: int64
child 3, contest_ceiling_usd: int64
child 4, cleanup_and_overhead_reserve_usd: int64
child 5, reference_gpu_rate_usd_hour: double
child 6, reference_source: string
child 7, job_execution_timeout_seconds: int64
child 8, admission_rule: string
child 9, retained_volume: string
child 10, note: string
child 11, authorized_by: string
child 12, allocation_revision: string
sha256: string
actual_cost_records: int64
file: string
estimated_cost_records: int64
scored_clips: int64
bytes: int64
to
{'file': Value('string'), 'sha256': Value('string'), 'bytes': Value('int64'), 'scored_clips': Value('int64'), 'actual_cost_records': Value('int64'), 'estimated_cost_records': Value('int64')}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
return get_rows(
dataset=dataset,
...<4 lines>...
column_names=column_names,
)
File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
return func(*args, **kwargs)
File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
yield from ds.decode(False) if ds.features else ds
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2951, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2461, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2486, in _iter_arrow
for key, pa_table in self.ex_iterable._iter_arrow():
~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 547, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 430, in _iter_arrow
for key, pa_table in self.generate_tables_fn(**gen_kwags):
~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
self._cast_table(pa_table, json_field_paths=json_field_paths),
~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
pa_table = table_cast(pa_table, self.info.features.arrow_schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
schema_version: string
version: string
format: string
format_authorized_by: string
paid_generation_authorized: bool
publication_authorized: bool
models: list<item: string>
child 0, item: string
volume: struct<id: string, region: string, models_root: string>
child 0, id: string
child 1, region: string
child 2, models_root: string
booths: list<item: struct<id: string, name: string, skill: string, setup: list<item: string>, action: list<i (... 1145 chars omitted)
child 0, item: struct<id: string, name: string, skill: string, setup: list<item: string>, action: list<item: string (... 1133 chars omitted)
child 0, id: string
child 1, name: string
child 2, skill: string
child 3, setup: list<item: string>
child 0, item: string
child 4, action: list<item: string>
child 0, item: string
child 5, ending: list<item: string>
child 0, item: string
child 6, continuity: list<item: string>
child 0, item: string
child 7, order: int64
child 8, seeds: list<item: int64>
child 0, item: int64
child 9, canonical_brief: string
child 10, prompt_variants: struct<ltx-2.5-distilled-bf16: struct<prompt: string, method: string, sources: list<item: string>, s (... 855 chars omitted)
child 0, ltx-2.5-distilled-bf16: struct<prompt: string, method: string, sources: list<item: string>, semantic_review: struct<type: st (... 191 chars omitted)
child 0, prompt: string
...
child 0, width: int64
child 1, height: int64
child 2, frames: int64
child 3, fps: string
child 2, wan2.2-t2v-a14b: struct<width: int64, height: int64, frames: int64, fps: string>
child 0, width: int64
child 1, height: int64
child 2, frames: int64
child 3, fps: string
child 3, image_digest: string
child 4, software_commits: struct<ltx-2.5-distilled-bf16: string, kandinsky-6-lite-distill-5s: string, wan2.2-t2v-a14b: string>
child 0, ltx-2.5-distilled-bf16: string
child 1, kandinsky-6-lite-distill-5s: string
child 2, wan2.2-t2v-a14b: string
child 5, endpoint_ids: null
child 6, manifest: string
child 7, manifest_sha256: string
spend_envelope: struct<authorized: bool, incremental_testing_ceiling_usd: int64, preflight_ceiling_usd: int64, conte (... 279 chars omitted)
child 0, authorized: bool
child 1, incremental_testing_ceiling_usd: int64
child 2, preflight_ceiling_usd: int64
child 3, contest_ceiling_usd: int64
child 4, cleanup_and_overhead_reserve_usd: int64
child 5, reference_gpu_rate_usd_hour: double
child 6, reference_source: string
child 7, job_execution_timeout_seconds: int64
child 8, admission_rule: string
child 9, retained_volume: string
child 10, note: string
child 11, authorized_by: string
child 12, allocation_revision: string
sha256: string
actual_cost_records: int64
file: string
estimated_cost_records: int64
scored_clips: int64
bytes: int64
to
{'file': Value('string'), 'sha256': Value('string'), 'bytes': Value('int64'), 'scored_clips': Value('int64'), 'actual_cost_records': Value('int64'), 'estimated_cost_records': Value('int64')}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
Reroll Rumble pilot — complete test record
How many attempts does an open-weight video model need for a usable clip? This Agentic Film Lab pilot runs LTX2.5 Distilledbf16, Wan2.2 T2V-A14B and Kandinsky6.0 Lite distilled5s through six frozen challenges. All36 scored native outputs are retained:17 accepted and19 rejected. One contestant reaches the eight-attempt cap in the finale. Results describe this recorded pilot, not universal model rankings.
Download the complete evidence archive, or browse the record. The ZIP contains all36 native MP4s, exact per-attempt requests/receipts/settings/seeds, frozen prompt variants and rules, creator judgments, runtime worker source and timing/cost provenance. The public record preserves historical review caveats and the creator’s later clarification about the accepted visible-top finale result.
Costs comprise28 actual bills and8 explicitly labeled creator-authorized GPU-execution estimates. Model scored-clip totals include passes and failures; CPU staging, storage, neutral checks and unscored administration are separate. Estimates do not fabricate settled bills and exclude unavailable startup/idle/extra-worker/disk/fee components. See record/model-cost-results.json and per-attempt receipts.
Footage is native and unchanged in the archive. The episode shows complete normal-speed scored clips and labeled previews/replays. Delivery is4K upscaled from the approved1080p composition, not native4K generation. Narration uses the consenting creator’s locally cloned voice; original series music was created locally with YuE2. No voice-reference recordings or unrelated private account/credential material are distributed.
The YouTube episode is initially uploaded privately at the creator’s direction; the evidence repository is public. Captions and allthree thumbnail alternatives are supplied here for the authorized upload and later thumbnail test. No full private episode master is included.
Weights are not included. Consult each model’s original license; this record does not replace those terms.
Official resources:
- https://huggingface.co/Lightricks/LTX-2.5
- https://huggingface.co/Wan-AI/Wan2.2-T2V-A14B
- https://huggingface.co/kandinskylab/Kandinsky-6.0-Lite-distill-5s-Diffusers
- https://github.com/Lightricks/LTX-2
- https://github.com/Wan-Video/Wan2.2
- https://huggingface.co/docs/diffusers/main/api/pipelines/kandinsky6
- https://docs.runpod.io/serverless/overview
- https://github.com/multimodal-art-projection/YuE
- https://github.com/resemble-ai/chatterbox
Archive SHA256: 129a1d6ef15056c3a4ecb2c4035c7d5a068d85a419e731c287eebafe97179b56
- Downloads last month
- 97