THE SOURCE IN CONTEXT
Four Spark expert workers are paired with one or two RTX PRO 6000 Blackwell coordinators, each limited to 400 W at standard memory clocks. The report identifies the base checkpoint, runtime source and image IDs. A 400 W RTX PRO 6000 is still that card; the experiment does not establish the behavior of a physical RTX 5090.
What is this hybrid system?
Four Spark expert workers are paired with one or two RTX PRO 6000 Blackwell coordinators, each limited to 400 W at standard memory clocks. The report identifies the base checkpoint, runtime source and image IDs. A 400 W RTX PRO 6000 is still that card; the experiment does not establish the behavior of a physical RTX 5090.
Source: tpurtell · pinned document
How was v14 compared with v13?
Each layout runs eight sessions in an interleaved version order, relaunching the appropriate image pair and discarding warmup. Sessions are the independent units. The published ratios use geometric means, with Welch 95% intervals on session log-ratios. The two-coordinator sampled C1 ratio of 1.377 is therefore a within-campaign version comparison, not a ratio against an unrelated desktop.
Source: tpurtell · pinned document
Why retain content and traffic type?
The release table’s weighted C1 medians are 104.69 tokens/s with one coordinator and 129.67 with two. Counting and code have different rates, and the concurrency sweep uses an aggregate interval from earliest output to final completion. That denominator differs from an individual request’s decode or end-to-end served rate.
Source: tpurtell · pinned document
What are the limitations and disclosed artifacts?
Some greedy outputs differ across arms through batch-shape numerics, although the author reports all objective checks passed. A schema-JSON probe initially hit the prompt cache and was rerun with a fresh nonce; the table uses that rerun. Historical startup and prefill tables belong to older campaigns. Keep the fresh decode evidence separate from those retained history sections.
Source: tpurtell · pinned document
Reported figures and their conditions
Selected facts reported by the original author. Each row has its own configuration and measurement boundary. Run AI Here has not reproduced these tests.
| Configuration / workload | Metric | Reported value | Evidence |
|---|---|---|---|
| One RTX coordinator + four Sparks, weighted DSpark | decode | 104.69 tokens/smedian of three repeats | Original row ↗ |
| Two RTX coordinators + four Sparks, weighted DSpark | decode | 129.67 tokens/smedian of three repeats | Original row ↗ |
| One RTX coordinator + four Sparks, target only | decode | 59.84 tokens/smedian of three repeats | Original row ↗ |
| Two RTX coordinators, sampled C1 v14/v13 | decode ratio | 1.377 [1.369–1.385] ratiogeometric mean of session-score ratio | Original row ↗ |
One RTX coordinator + four Sparks, weighted DSpark · decode · conditions
| system | 1 or 2 RTX PRO 6000 Blackwell coordinators plus 4 DGX Spark workers |
|---|---|
| node_count | Not specified |
| parallelism | 4 Spark expert workers, TP4 |
| kv_dtype | Not specified |
| context_limit_tokens | Not specified |
| experiment_date | 2026-09-24 |
| model_revision | dba1be0a40aa45a94ad051997016db3960a90277 |
| runtime_execution_revision | af9d96923f70af1cd385960e28450e5ec0653ff8 |
| worker_count | 4 |
| worker_parallelism | Spark TP4 |
| gpu_power_watts_each | 400 |
| repetitions | 3 |
| statistic | median |
| concurrency | 1 |
| coordinator_count | 1 |
Two RTX coordinators + four Sparks, weighted DSpark · decode · conditions
| system | 1 or 2 RTX PRO 6000 Blackwell coordinators plus 4 DGX Spark workers |
|---|---|
| node_count | Not specified |
| parallelism | 4 Spark expert workers, TP4 |
| kv_dtype | Not specified |
| context_limit_tokens | Not specified |
| experiment_date | 2026-09-24 |
| model_revision | dba1be0a40aa45a94ad051997016db3960a90277 |
| runtime_execution_revision | af9d96923f70af1cd385960e28450e5ec0653ff8 |
| worker_count | 4 |
| worker_parallelism | Spark TP4 |
| gpu_power_watts_each | 400 |
| repetitions | 3 |
| statistic | median |
| concurrency | 1 |
| coordinator_count | 2 |
One RTX coordinator + four Sparks, target only · decode · conditions
| system | 1 or 2 RTX PRO 6000 Blackwell coordinators plus 4 DGX Spark workers |
|---|---|
| node_count | Not specified |
| parallelism | 4 Spark expert workers, TP4 |
| kv_dtype | Not specified |
| context_limit_tokens | Not specified |
| experiment_date | 2026-09-24 |
| model_revision | dba1be0a40aa45a94ad051997016db3960a90277 |
| runtime_execution_revision | af9d96923f70af1cd385960e28450e5ec0653ff8 |
| worker_count | 4 |
| worker_parallelism | Spark TP4 |
| gpu_power_watts_each | 400 |
| repetitions | 3 |
| statistic | median |
| concurrency | 1 |
| coordinator_count | 1 |
| speculative_decoding | False |
Two RTX coordinators, sampled C1 v14/v13 · decode ratio · conditions
| system | 1 or 2 RTX PRO 6000 Blackwell coordinators plus 4 DGX Spark workers |
|---|---|
| node_count | Not specified |
| parallelism | 4 Spark expert workers, TP4 |
| kv_dtype | Not specified |
| context_limit_tokens | Not specified |
| experiment_date | 2026-09-24 |
| model_revision | dba1be0a40aa45a94ad051997016db3960a90277 |
| runtime_execution_revision | af9d96923f70af1cd385960e28450e5ec0653ff8 |
| worker_count | 4 |
| worker_parallelism | Spark TP4 |
| gpu_power_watts_each | 400 |
| repetitions | 8 interleaved sessions per layout; four per arm |
| statistic | median |
| concurrency | 1 |
| coordinator_count | 2 |
| temperature | 0.7 |
| top_p | 0.9 |
{'interval_type': 'Welch 95% interval on session log-ratios', 'lower': 1.369, 'upper': 1.385}
AI-assisted source reading and original editorial synthesis; no independent model execution.