Datasets:

orwelian84
/

Qwen3.8-27B-Three-GPU-Benchmarks

Tags:

Dataset Viewer

Split (1)

train

The dataset viewer is not available for this split.

Cannot load the dataset split (in streaming mode) to extract the first rows.

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
cards: list<item: struct<card: string, evidence_ids: list<item: string>, generated_tokens_per_request: list (... 1467 chars omitted)
  child 0, item: struct<card: string, evidence_ids: list<item: string>, generated_tokens_per_request: list<item: int6 (... 1455 chars omitted)
      child 0, card: string
      child 1, evidence_ids: list<item: string>
          child 0, item: string
      child 2, generated_tokens_per_request: list<item: int64>
          child 0, item: int64
      child 3, models: list<item: struct<bytes: int64, name: string, quantization: string, sha256: string>>
          child 0, item: struct<bytes: int64, name: string, quantization: string, sha256: string>
              child 0, bytes: int64
              child 1, name: string
              child 2, quantization: string
              child 3, sha256: string
      child 4, profiles: list<item: struct<completed_runs: int64, config: struct<batch: int64, context: int64, flash_attentio (... 1169 chars omitted)
          child 0, item: struct<completed_runs: int64, config: struct<batch: int64, context: int64, flash_attention: string,  (... 1157 chars omitted)
              child 0, completed_runs: int64
              child 1, config: struct<batch: int64, context: int64, flash_attention: string, kv_k: string, kv_v: string, parallel:  (... 216 chars omitted)
                  child 0, batch: int64
                  child 1, context: int64
                  child 2, flash_attention: string
                  child 3,
...
 card: string
      child 1, baseline: struct<context: int64, parallel: int64, batch: int64, ubatch: int64, kv_k: string, kv_v: string, fla (... 21 chars omitted)
          child 0, context: int64
          child 1, parallel: int64
          child 2, batch: int64
          child 3, ubatch: int64
          child 4, kv_k: string
          child 5, kv_v: string
          child 6, flash_attention: string
      child 2, mtp: struct<context: int64, parallel: int64, batch: int64, ubatch: int64, kv_k: string, kv_v: string, fla (... 64 chars omitted)
          child 0, context: int64
          child 1, parallel: int64
          child 2, batch: int64
          child 3, ubatch: int64
          child 4, kv_k: string
          child 5, kv_v: string
          child 6, flash_attention: string
          child 7, draft_max: int64
          child 8, draft_gpu_layers: int64
      child 3, repeats: int64
llama_cpp_revision: string
model_revision: string
concurrent_profiles: list<item: struct<card: string, context: int64, parallel: int64, slot_context: int64, batch: int64,  (... 59 chars omitted)
  child 0, item: struct<card: string, context: int64, parallel: int64, slot_context: int64, batch: int64, ubatch: int (... 47 chars omitted)
      child 0, card: string
      child 1, context: int64
      child 2, parallel: int64
      child 3, slot_context: int64
      child 4, batch: int64
      child 5, ubatch: int64
      child 6, kv_k: string
      child 7, kv_v: string
      child 8, repeats: int64
to
{'schema': Value('string'), 'llama_cpp_revision': Value('string'), 'model_revision': Value('string'), 'decode': {'n_predict': Value('int64'), 'temperature': Value('float64'), 'top_k': Value('int64'), 'top_p': Value('float64'), 'ignore_eos': Value('bool'), 'cache_prompt': Value('bool'), 'stream': Value('bool')}, 'common_server_args': {'gpu_layers': Value('int64'), 'threads': Value('int64'), 'threads_batch': Value('int64'), 'metrics': Value('bool'), 'webui': Value('bool')}, 'interactive_pairs': List({'card': Value('string'), 'baseline': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string')}, 'mtp': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string'), 'draft_max': Value('int64'), 'draft_gpu_layers': Value('int64')}, 'repeats': Value('int64')}), 'ngram_profiles': {'cards': List(Value('string')), 'common': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string'), 'repeats': Value('int64')}, 'profiles': List({'id': Value('string'), 'spec_type': Value('string'), 'draft_max': Value('int64'), 'ngram_n': Value('int64'), 'ngram_m': Value('int64'), 'ngram_min_hits': Value('int64')})}, 'concurrent_profiles': List({'card': Value('string'), 'context': Value('int64'), 'parallel': Value('int64'), 'slot_context': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'repeats': 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 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, 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 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, 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
              cards: list<item: struct<card: string, evidence_ids: list<item: string>, generated_tokens_per_request: list (... 1467 chars omitted)
                child 0, item: struct<card: string, evidence_ids: list<item: string>, generated_tokens_per_request: list<item: int6 (... 1455 chars omitted)
                    child 0, card: string
                    child 1, evidence_ids: list<item: string>
                        child 0, item: string
                    child 2, generated_tokens_per_request: list<item: int64>
                        child 0, item: int64
                    child 3, models: list<item: struct<bytes: int64, name: string, quantization: string, sha256: string>>
                        child 0, item: struct<bytes: int64, name: string, quantization: string, sha256: string>
                            child 0, bytes: int64
                            child 1, name: string
                            child 2, quantization: string
                            child 3, sha256: string
                    child 4, profiles: list<item: struct<completed_runs: int64, config: struct<batch: int64, context: int64, flash_attentio (... 1169 chars omitted)
                        child 0, item: struct<completed_runs: int64, config: struct<batch: int64, context: int64, flash_attention: string,  (... 1157 chars omitted)
                            child 0, completed_runs: int64
                            child 1, config: struct<batch: int64, context: int64, flash_attention: string, kv_k: string, kv_v: string, parallel:  (... 216 chars omitted)
                                child 0, batch: int64
                                child 1, context: int64
                                child 2, flash_attention: string
                                child 3,
              ...
               card: string
                    child 1, baseline: struct<context: int64, parallel: int64, batch: int64, ubatch: int64, kv_k: string, kv_v: string, fla (... 21 chars omitted)
                        child 0, context: int64
                        child 1, parallel: int64
                        child 2, batch: int64
                        child 3, ubatch: int64
                        child 4, kv_k: string
                        child 5, kv_v: string
                        child 6, flash_attention: string
                    child 2, mtp: struct<context: int64, parallel: int64, batch: int64, ubatch: int64, kv_k: string, kv_v: string, fla (... 64 chars omitted)
                        child 0, context: int64
                        child 1, parallel: int64
                        child 2, batch: int64
                        child 3, ubatch: int64
                        child 4, kv_k: string
                        child 5, kv_v: string
                        child 6, flash_attention: string
                        child 7, draft_max: int64
                        child 8, draft_gpu_layers: int64
                    child 3, repeats: int64
              llama_cpp_revision: string
              model_revision: string
              concurrent_profiles: list<item: struct<card: string, context: int64, parallel: int64, slot_context: int64, batch: int64,  (... 59 chars omitted)
                child 0, item: struct<card: string, context: int64, parallel: int64, slot_context: int64, batch: int64, ubatch: int (... 47 chars omitted)
                    child 0, card: string
                    child 1, context: int64
                    child 2, parallel: int64
                    child 3, slot_context: int64
                    child 4, batch: int64
                    child 5, ubatch: int64
                    child 6, kv_k: string
                    child 7, kv_v: string
                    child 8, repeats: int64
              to
              {'schema': Value('string'), 'llama_cpp_revision': Value('string'), 'model_revision': Value('string'), 'decode': {'n_predict': Value('int64'), 'temperature': Value('float64'), 'top_k': Value('int64'), 'top_p': Value('float64'), 'ignore_eos': Value('bool'), 'cache_prompt': Value('bool'), 'stream': Value('bool')}, 'common_server_args': {'gpu_layers': Value('int64'), 'threads': Value('int64'), 'threads_batch': Value('int64'), 'metrics': Value('bool'), 'webui': Value('bool')}, 'interactive_pairs': List({'card': Value('string'), 'baseline': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string')}, 'mtp': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string'), 'draft_max': Value('int64'), 'draft_gpu_layers': Value('int64')}, 'repeats': Value('int64')}), 'ngram_profiles': {'cards': List(Value('string')), 'common': {'context': Value('int64'), 'parallel': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'flash_attention': Value('string'), 'repeats': Value('int64')}, 'profiles': List({'id': Value('string'), 'spec_type': Value('string'), 'draft_max': Value('int64'), 'ngram_n': Value('int64'), 'ngram_m': Value('int64'), 'ngram_min_hits': Value('int64')})}, 'concurrent_profiles': List({'card': Value('string'), 'context': Value('int64'), 'parallel': Value('int64'), 'slot_context': Value('int64'), 'batch': Value('int64'), 'ubatch': Value('int64'), 'kv_k': Value('string'), 'kv_v': Value('string'), 'repeats': Value('int64')})}
              because column names don't match

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.

Qwen3.8-27B three-GPU llama.cpp benchmarks

This repository contains the public reproduction package for a local Qwen3.8-27B inference study. The study covers an RTX 3090 24 GB, an RTX 4090 24 GB, and an RTX PRO 6000 Blackwell 96 GB.

Only repeated runs that passed an independent artifact validator appear in confirmed-results.csv. Single-run tuning points and invalidated attempts are not headline results.

ARTICLE.md contains the living Hugging Face article source.
confirmed-results.csv contains repeated medians and observed ranges.
profiles.json contains the exact public runtime profiles.
prompts.md contains the exact synthetic prompts and request settings.
public/NGRAM.md contains the matched five-repeat prompt n-gram control.
public/ contains the scrubbed expanded reports and charts.

Pinned inputs

llama.cpp commit: 9b05354ec6fb58b4e665e9a39ebc40285c015638
GGUF repository: ggml-org/Qwen3.8-27B-GGUF
GGUF revision: 0669b98607d47046c7c2b3f801011d54a08cfccf
Q4_K_M SHA-256: 31629f53165ab6a7dad8c9847dcfd1fdf55829dac1e6e748f4a68581b0033d34
Q4_0 MTP SHA-256: 051a1764cff8c4f3ee6ae8b00593a0364c7539c67fa50ffc58f3f96509fca38e
Q8_0 SHA-256: f5c702d8820d36fb55985bb238fc83ee3a313e920f4b752a437c3a6a9e14e4c8
Official BF16 model revision: 1d4bf0f2ff6012fd82039f2fa52739d0dd7c60c0
Locally converted official BF16 GGUF SHA-256: 93f653bed5ffa1993caefa0d1638b5dcabc1af0fd15c879199e5f72d048b6c0d

Measurement rule

Each interactive row generates 512 tokens from the same astronomy prompt. The table reports the median and observed range. Baseline and MTP profiles are matched within each card. Card-specific cache and batch settings prevent a strict card-to-card hardware comparison.

Synthetic prompt throughput does not measure model quality. Q4 and Q8 MTP profiles produced stable repeated outputs, but those outputs were not byte-identical to their non-speculative baselines.

Prompt n-gram results are also reported only after five matched repeats. The effect varied by prompt and card. The article does not present prompt n-gram decoding as a global default.

Reproduction

Build llama.cpp at the pinned commit with CUDA enabled.
Download the pinned model revision and verify each SHA-256 value.
Start llama-server with one profile from profiles.json.
Submit the payloads in prompts.md with greedy decoding.
Run the stated repeat count and report the median.

The expanded method and platform details are in public/REPRODUCIBILITY.md.

Downloads last month: 70

Total file size:

329 kB

Qwen3.8-27B three-GPU llama.cpp benchmarks

Contents

Pinned inputs

Measurement rule

Reproduction