Hugging Face's logo

Datasets:
livadies
/
t4-proof-runs

Tasks:
Image-Text-to-Text
Automatic Speech Recognition
Text-to-Speech
Languages:
English
Size:
n<1K
Tags:
benchmarking
kaggle
tesla-t4
reproducibility
open-models
License:
Dataset card Data Studio Files
xet
 Community
Dataset Preview
Duplicate
The full dataset viewer is not available (click to read why). Only showing a preview of the rows.
The dataset generation failed
Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
$schema: string
$id: string
title: string
type: string
required: list<item: string>
  child 0, item: string
properties: struct<id: struct<type: string, pattern: string>, model: struct<type: string>, publisher: struct<typ (... 661 chars omitted)
  child 0, id: struct<type: string, pattern: string>
      child 0, type: string
      child 1, pattern: string
  child 1, model: struct<type: string>
      child 0, type: string
  child 2, publisher: struct<type: string>
      child 0, type: string
  child 3, task: struct<enum: list<item: string>>
      child 0, enum: list<item: string>
          child 0, item: string
  child 4, status: struct<enum: list<item: string>>
      child 0, enum: list<item: string>
          child 0, item: string
  child 5, verdict: struct<type: string>
      child 0, type: string
  child 6, license: struct<type: string>
      child 0, type: string
  child 7, commercial_use: struct<type: string>
      child 0, type: string
  child 8, weights_gib: struct<type: list<item: string>, minimum: int64>
      child 0, type: list<item: string>
          child 0, item: string
      child 1, minimum: int64
  child 9, measured_vram_gib: struct<type: list<item: string>, minimum: int64>
      child 0, type: list<item: string>
          child 0, item: string
      child 1, minimum: int64
  child 10, cold_start_seconds: struct<type: list<item: string>, minimum: int64>
      child 0, type: list<item: string>
          child 0, item: string
      child 1, minimum: int64
  child 11, precision: struct<type: string>
      child 0, type: string
  child 12, hardware: struct<type: string>
      child 0, type: string
  child 13, source_url: struct<type: string, format: string>
      child 0, type: string
      child 1, format: string
  child 14, artifact_url: struct<type: list<item: string>, format: string>
      child 0, type: list<item: string>
          child 0, item: string
      child 1, format: string
  child 15, run_at: struct<type: list<item: string>, format: string>
      child 0, type: list<item: string>
          child 0, item: string
      child 1, format: string
  child 16, notes: struct<type: string>
      child 0, type: string
additionalProperties: bool
to
{'id': Value('string'), 'model': Value('string'), 'publisher': Value('string'), 'task': Value('string'), 'status': Value('string'), 'verdict': Value('string'), 'license': Value('string'), 'commercial_use': Value('bool'), 'weights_gib': Value('float64'), 'measured_vram_gib': Value('null'), 'cold_start_seconds': Value('null'), 'precision': Value('string'), 'hardware': Value('string'), 'trend_rank': Value('int64'), 'source_url': Value('string'), 'artifact_url': Value('null'), 'run_at': Value('null'), 'notes': Value('string'), 'downloads': Value('int64'), 'likes': Value('int64'), 'last_modified': Value('string'), 'pipeline_tag': Value('string'), 'parameter_count': Value('int64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1858, in _prepare_split_single
                  num_examples, num_bytes = writer.finalize()
                                            ~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 781, in finalize
                  self.write_rows_on_file()
                  ~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 663, in write_rows_on_file
                  self._write_table(table)
                  ~~~~~~~~~~~~~~~~~^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/arrow_writer.py", line 773, in _write_table
                  pa_table = table_cast(pa_table, self._schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2369, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              $schema: string
              $id: string
              title: string
              type: string
              required: list<item: string>
                child 0, item: string
              properties: struct<id: struct<type: string, pattern: string>, model: struct<type: string>, publisher: struct<typ (... 661 chars omitted)
                child 0, id: struct<type: string, pattern: string>
                    child 0, type: string
                    child 1, pattern: string
                child 1, model: struct<type: string>
                    child 0, type: string
                child 2, publisher: struct<type: string>
                    child 0, type: string
                child 3, task: struct<enum: list<item: string>>
                    child 0, enum: list<item: string>
                        child 0, item: string
                child 4, status: struct<enum: list<item: string>>
                    child 0, enum: list<item: string>
                        child 0, item: string
                child 5, verdict: struct<type: string>
                    child 0, type: string
                child 6, license: struct<type: string>
                    child 0, type: string
                child 7, commercial_use: struct<type: string>
                    child 0, type: string
                child 8, weights_gib: struct<type: list<item: string>, minimum: int64>
                    child 0, type: list<item: string>
                        child 0, item: string
                    child 1, minimum: int64
                child 9, measured_vram_gib: struct<type: list<item: string>, minimum: int64>
                    child 0, type: list<item: string>
                        child 0, item: string
                    child 1, minimum: int64
                child 10, cold_start_seconds: struct<type: list<item: string>, minimum: int64>
                    child 0, type: list<item: string>
                        child 0, item: string
                    child 1, minimum: int64
                child 11, precision: struct<type: string>
                    child 0, type: string
                child 12, hardware: struct<type: string>
                    child 0, type: string
                child 13, source_url: struct<type: string, format: string>
                    child 0, type: string
                    child 1, format: string
                child 14, artifact_url: struct<type: list<item: string>, format: string>
                    child 0, type: list<item: string>
                        child 0, item: string
                    child 1, format: string
                child 15, run_at: struct<type: list<item: string>, format: string>
                    child 0, type: list<item: string>
                        child 0, item: string
                    child 1, format: string
                child 16, notes: struct<type: string>
                    child 0, type: string
              additionalProperties: bool
              to
              {'id': Value('string'), 'model': Value('string'), 'publisher': Value('string'), 'task': Value('string'), 'status': Value('string'), 'verdict': Value('string'), 'license': Value('string'), 'commercial_use': Value('bool'), 'weights_gib': Value('float64'), 'measured_vram_gib': Value('null'), 'cold_start_seconds': Value('null'), 'precision': Value('string'), 'hardware': Value('string'), 'trend_rank': Value('int64'), 'source_url': Value('string'), 'artifact_url': Value('null'), 'run_at': Value('null'), 'notes': Value('string'), 'downloads': Value('int64'), 'likes': Value('int64'), 'last_modified': Value('string'), 'pipeline_tag': Value('string'), 'parameter_count': Value('int64')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1683, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1869, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

id
string
model
string
publisher
string
task
string
status
string
verdict
string
license
string
commercial_use
bool
weights_gib
float64
measured_vram_gib
null
cold_start_seconds
null
precision
string
hardware
string
trend_rank
int64
source_url
string
artifact_url
null
run_at
null
notes
string
downloads
int64
likes
int64
last_modified
string
pipeline_tag
string
parameter_count
int64
abot-world-0-5b-lf
ABot-World-0.5B-LF
ACVLab
Video
running
UNDER TEST
Apache-2.0
true
9.82
null
null
FP16
2× Tesla T4
5
https://huggingface.co/acvlab/ABot-World-0-5B-LF
null
null
A 0.5B interactive-world model with a heavy video stack. The first dual-T4 compatibility rollout is currently in the Kaggle queue.
191
21
2026-07-13T05:06:34.000Z
image-to-video
null
unlimited-ocr
Unlimited-OCR
Baidu
Vision
candidate
QUEUED
MIT
true
6.21
null
null
BF16/FP16
2× Tesla T4
1
https://huggingface.co/baidu/Unlimited-OCR
null
null
A high-interest OCR model and a practical first vision target: small enough to be plausible, useful enough to make the result matter.
1,506,937
1,963
2026-07-03T05:15:09.000Z
image-text-to-text
3,336,106,240
moss-transcribe-diarize
MOSS-Transcribe-Diarize
OpenMOSS Team
Audio
candidate
QUEUED
Apache-2.0
true
1.69
null
null
FP16
2× Tesla T4
2
https://huggingface.co/OpenMOSS-Team/MOSS-Transcribe-Diarize
null
null
Transcription plus speaker diarization in one testable workflow. The proof artifact will be timestamped speaker turns and runtime telemetry.
39,509
160
2026-07-12T07:03:41.000Z
audio-text-to-text
908,513,280
gepard-1-0
Gepard 1.0
NineNineSix
Audio
candidate
QUEUED
Apache-2.0
true
1.04
null
null
FP16
2× Tesla T4
3
https://huggingface.co/nineninesix/gepard-1.0
null
null
A compact speech model candidate. We will publish both an audio sample and real-time factor, not just a successful import.
3,940
95
2026-07-11T12:04:14.000Z
text-to-speech
555,694,169
supra-router-51m
Supra Router 51M
SupraLabs
Language
candidate
LICENSE CHECK
Review needed
false
0.1
null
null
FP16
1× Tesla T4
4
https://huggingface.co/SupraLabs/Supra-Router-51M
null
null
Tiny enough for an interesting routing-speed baseline, but it stays outside the commercial filter until its model card license is explicit.
1,573
113
2026-07-09T20:38:41.000Z
text-generation
51,786,240
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null
null

T4 Proof — Open Model Reality Check

Not estimated. Actually run. This dataset tracks reproducible runs of trending open models on Kaggle's free Dual Tesla T4 environment.

What makes a run verified?

A row may be a candidate, running, failed, or verified. Verified is reserved for a run that has all of the following:

  1. a recorded Kaggle hardware and software environment;
  2. a fixed, representative input and generation configuration;
  3. cold-start, peak-VRAM, and task timing measurements;
  4. an inspectable output artifact;
  5. logs or a notebook sufficient to reproduce the result.

A successful model import alone is not a verified task run.

Files

  • data/benchmarks.json — dashboard records and current queue.
  • schema/benchmark-result.schema.json — validation contract.
  • README.md — this dataset card.

Hardware baseline

The initial environment probe recorded two Tesla T4 GPUs with 14.56 GiB usable VRAM each, CUDA 12.8, and PyTorch 2.10.

Live dashboard

The interactive board is available at T4 Proof.

Limitations

The dataset starts intentionally small. Candidate status is not a performance claim. Results are specific to the recorded revision, precision, runtime, and task adapter; they should not be generalized to different model revisions or hardware.

License

Dataset metadata and benchmark harness: Apache-2.0. Individual model licenses remain with their respective publishers and are recorded per row.
Downloads last month
-
Total file size:
9.86 kB