# DeepSeek 1017 on Two Sparks: Read the TensorFold Evidence

There is now a community report that explicitly names DeepSeek-V4.1-Flash-1017 on a pair of DGX Sparks. Its useful contribution is a traceable model view, saved results and documented failure boundaries. Treat it as evidence for those particular configurations. It does not establish an engine-wide speed advantage or general model quality.

This note reads [Elevatormusic’s pinned recipe](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/README.md), including the October 17 comparison and the separate October 18 follow-up. Run AI Here has not run this model or reproduced the author’s results.

![Selected 1017 source decode observations and separate final-setup prefill medians with five original row values.](/assets/deepseek-1017-source-conditions.webp)

Figure: drawn directly from selected source values. The two decode arms and the later prefill cohort are separate. Prefill whiskers show an observed minimum and maximum, not a confidence interval.

## First verify which 1017 files were served

The author combines `LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw` at `afc7aa0aab98ee735c0f54e0a122019af1ac08f1` with two Engram shards from `deepseek-ai/DeepSeek-V4.1-Flash-1017` at `1bae0e1088993b9b82de08f59de5be5a34e42170`. This identifies an EXL3 conversion and its separate base-model tables. It is not a measurement of the unconverted official checkpoint. [Model-view provenance](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/PROVENANCE.md#L7).

The pinned limits document says that the measured conversion revision was already behind a newer card revision. The author did not compare them. Keep the measured revision when reusing a number; a current download is not automatically the same artifact.

## Keep one named build beside each decode value

These selected observations use one request, thinking off and two prompts per workload. The suite streams with `max_tokens=700`, temperature 1.0 and top-p 1.0. Actual output lengths and exact prompt-token counts are not supplied in the selected result rows. Decode excludes the first-token wait: `(completion_tokens − 1) / (stream end − first content-bearing chunk)`.

| October 17 arm | Prose decode, tokens/s | Code decode, tokens/s |
|---|---:|---:|
| vLLM kit, run 01 | 27.0 | 39.9 |
| TensorFold, run 11 | 56.9 | 91.8 |

Each cell is the source’s rounded mean of two prompt-level rates within one run. It has no supplied repeated-run uncertainty. [Original vLLM result](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/01_vllm-miaai-kit_suite-results_32of32.json), [original TensorFold result](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/11_tensorfold-fd2365042f56_suite-results_32of32.json).

The report’s headline median combines six different TensorFold builds. It is not six repetitions of this run-11 configuration. The selected values above deliberately identify one build instead of hiding that distinction.

## The comparison changes more than the engine

Both arms use the same reported model view and Spark pair, but their serving settings differ. The vLLM kit uses an exllamav3 overlay, three draft tokens, one configured sequence, a 262,144-token cap and 1,536-token batching. TensorFold run 11 uses Python 0.6.6 at `fd2365042f56fcf726cf4071132c8b779ad2eae0`, five drafts, capacity for four requests, a 1,048,576-token cap and 2,048-token prefill chunks. The source also records a template patch, with its per-run use inferred from file times. [Comparison conditions](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/docs/BENCHMARKS.md#L32).

The baseline image digest is retained in JSON. The TensorFold executed image digest is unknown. The recipe now builds a later `92aa710b533d` revision prefix. Do not attach the October 17 decode cells to that later build or describe this as an isolated engine-only experiment. No multi-user vLLM comparison was measured.

## The newer long-prompt cohort answers a different question

The October 18 final setup uses random-word prompts and records zero cached tokens. Five source rows give:

| Prompt bucket | Median prefill, tokens/s | Observed range |
|---|---:|---:|
| 32K | 2,165 | 2,065–2,201 |
| 128K | 1,961 | 1,851–1,986 |

The denominator is prompt tokens divided by TTFT. These runs include a defined warm-up sequence, Engram page handling and host memory controls. They are separate from the older suite’s repeated-sentence prompts, where prefix reuse could not be ruled out. Each original row remains in the export. [Final setup and five rows](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-18/INDEX.md#L5).

For deployment planning, inspect per-node disk, runtime cache and memory headroom separately. The author requires a substantial local model view and additional first-start cache. Host tuning affects the whole machine: read the original settings and their measured scope before adopting them. This article does not execute or prescribe them for your cluster.

## What the tests still leave open

The author reports two near-million-token needle tests, with about 1.7 GiB available on the head at the memory low point. A retrieved fact does not validate general long-context reasoning, near-limit concurrent sessions or decode speed. Image input was not tested. Neither the task suite nor the weights’ separate perplexity claim establishes this engine’s general accuracy. Some slower states remain unexplained. [Known limits](https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/docs/KNOWN-LIMITS.md#L6).

Use this recipe as a concrete reproduction candidate. Preserve the model view, exact arm, template, draft depth, cache state and request shape before judging a local result. The [JSON source record](/community/deepseek-1017-tensorfold-spark/index.json) and adjacent article exports keep those fields with the six selected measurements.


## Original sources and conditions

- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/README.md
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: 546a0c3bfbbb6e84226ec812cf38927f6d993e53b73633a540c426eea5f13d29
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/docs/BENCHMARKS.md
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: b2d05a077ae8145a2cf2a98ac2ae2eabef87b992230c32398a60e0a2e9174861
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/docs/KNOWN-LIMITS.md
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: eeb6bc6f7a3fce20654f977581e9718b124e70fc01ab358efebefe9afd9dc942
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/PROVENANCE.md
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: aa020bd6aa0fe840c9518c96dbc00f04bbdf6add834f2fe7bac2b076aa91a089
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-18/INDEX.md
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: e69d8c270e6407ddbaacb82da36cc48994b6f20fd4300d176753371b4e23fb8e
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/01_vllm-miaai-kit_suite-results_32of32.json
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: 0cb7c61b5b321ba0b9eec76460e4d9d3d0e759bd09cfb03fc747ef66e7c2ea63
- Elevatormusic: https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/11_tensorfold-fd2365042f56_suite-results_32of32.json
  - Document revision: 9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32
  - SHA-256: d0ec8c99b0208ac4ddc272379d3cd71c796b5662d09481d6ba1e98fbbf0fa41c
- deepseek-ai: https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash-1017/blob/1bae0e1088993b9b82de08f59de5be5a34e42170/README.md
  - Document revision: 1bae0e1088993b9b82de08f59de5be5a34e42170
  - SHA-256: b72b796e4f4c739eca5d0ad360a7679b39118ca7eb64891f7151a527b292c42a

Checked 2026-10-18. AI-assisted writing, source checked by Run AI Here. No independent model tests. Conditions and selected source facts are in the adjacent JSON export.

```json
{
  "facts": [],
  "reported_results": [
    {
      "label": "vllm-kit-run01 prose",
      "metric": "decode_rate",
      "value": 27.0,
      "unit": "tokens/s",
      "statistic": "one run; mean of two prompt-level decode rates; rounded by source",
      "uncertainty": null,
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-17",
        "concurrency": 1,
        "thinking": false,
        "temperature": 1.0,
        "top_p": 1.0,
        "max_output_tokens": 700,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": "present; cache-hit counts not logged in this suite",
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "(completion_tokens - 1) / (stream end - first chunk with content, reasoning or tool call)",
        "repetitions": 1,
        "checkpoint_is_latest_revision_verified": false,
        "id": "vllm-kit-run01",
        "engine": "vLLM + exllamav3 overlay",
        "engine_version": null,
        "runtime_execution_revision": null,
        "image_digest": "sha256:2f0cf3adc0f989c1d446be274df864eb799630175f604c3b22b71b7205971dce",
        "kit_revision": "6f7d1590ad49a2b8995188e45d7b9db31e677452",
        "draft_tokens": 3,
        "configured_max_requests": 1,
        "context_limit_tokens": 262144,
        "prefill_chunk_tokens": 1536,
        "chat_template": "kit template; actual CHAT_TEMPLATE override not read by the source",
        "workload": "prose",
        "per_prompt_decode_rates": [
          24.7,
          29.3
        ],
        "per_prompt_TTFT_seconds": [
          0.32,
          0.28
        ],
        "prompt_count": 2
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/01_vllm-miaai-kit_suite-results_32of32.json#L6",
      "source_document_sha256": "0cb7c61b5b321ba0b9eec76460e4d9d3d0e759bd09cfb03fc747ef66e7c2ea63",
      "source_document_id": "ds1017-runs-2026-10-17-01_vllm-miaai-kit_suite-results_32of32.json",
      "source_line": 6,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    },
    {
      "label": "vllm-kit-run01 code",
      "metric": "decode_rate",
      "value": 39.9,
      "unit": "tokens/s",
      "statistic": "one run; mean of two prompt-level decode rates; rounded by source",
      "uncertainty": null,
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-17",
        "concurrency": 1,
        "thinking": false,
        "temperature": 1.0,
        "top_p": 1.0,
        "max_output_tokens": 700,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": "present; cache-hit counts not logged in this suite",
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "(completion_tokens - 1) / (stream end - first chunk with content, reasoning or tool call)",
        "repetitions": 1,
        "checkpoint_is_latest_revision_verified": false,
        "id": "vllm-kit-run01",
        "engine": "vLLM + exllamav3 overlay",
        "engine_version": null,
        "runtime_execution_revision": null,
        "image_digest": "sha256:2f0cf3adc0f989c1d446be274df864eb799630175f604c3b22b71b7205971dce",
        "kit_revision": "6f7d1590ad49a2b8995188e45d7b9db31e677452",
        "draft_tokens": 3,
        "configured_max_requests": 1,
        "context_limit_tokens": 262144,
        "prefill_chunk_tokens": 1536,
        "chat_template": "kit template; actual CHAT_TEMPLATE override not read by the source",
        "workload": "code",
        "per_prompt_decode_rates": [
          40.4,
          39.4
        ],
        "per_prompt_TTFT_seconds": [
          0.47,
          0.32
        ],
        "prompt_count": 2
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/01_vllm-miaai-kit_suite-results_32of32.json#L20",
      "source_document_sha256": "0cb7c61b5b321ba0b9eec76460e4d9d3d0e759bd09cfb03fc747ef66e7c2ea63",
      "source_document_id": "ds1017-runs-2026-10-17-01_vllm-miaai-kit_suite-results_32of32.json",
      "source_line": 20,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    },
    {
      "label": "tensorfold-run11 prose",
      "metric": "decode_rate",
      "value": 56.9,
      "unit": "tokens/s",
      "statistic": "one run; mean of two prompt-level decode rates; rounded by source",
      "uncertainty": null,
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-17",
        "concurrency": 1,
        "thinking": false,
        "temperature": 1.0,
        "top_p": 1.0,
        "max_output_tokens": 700,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": "present; cache-hit counts not logged in this suite",
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "(completion_tokens - 1) / (stream end - first chunk with content, reasoning or tool call)",
        "repetitions": 1,
        "checkpoint_is_latest_revision_verified": false,
        "id": "tensorfold-run11",
        "engine": "TensorFold",
        "engine_version": "0.6.6",
        "runtime_execution_revision": "fd2365042f56fcf726cf4071132c8b779ad2eae0",
        "image_tag": "tensorfold-dsv41:fd2365042f56",
        "image_digest": null,
        "draft_tokens": 5,
        "configured_max_requests": 4,
        "context_limit_tokens": 1048576,
        "prefill_chunk_tokens": 2048,
        "chat_template": "patched kit template; source infers per-run use from file times",
        "workload": "prose",
        "per_prompt_decode_rates": [
          53.4,
          60.4
        ],
        "per_prompt_TTFT_seconds": [
          0.12,
          0.12
        ],
        "prompt_count": 2
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/11_tensorfold-fd2365042f56_suite-results_32of32.json#L6",
      "source_document_sha256": "d0ec8c99b0208ac4ddc272379d3cd71c796b5662d09481d6ba1e98fbbf0fa41c",
      "source_document_id": "ds1017-runs-2026-10-17-11_tensorfold-fd2365042f56_suite-results_32of32.json",
      "source_line": 6,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    },
    {
      "label": "tensorfold-run11 code",
      "metric": "decode_rate",
      "value": 91.8,
      "unit": "tokens/s",
      "statistic": "one run; mean of two prompt-level decode rates; rounded by source",
      "uncertainty": null,
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-17",
        "concurrency": 1,
        "thinking": false,
        "temperature": 1.0,
        "top_p": 1.0,
        "max_output_tokens": 700,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": "present; cache-hit counts not logged in this suite",
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "(completion_tokens - 1) / (stream end - first chunk with content, reasoning or tool call)",
        "repetitions": 1,
        "checkpoint_is_latest_revision_verified": false,
        "id": "tensorfold-run11",
        "engine": "TensorFold",
        "engine_version": "0.6.6",
        "runtime_execution_revision": "fd2365042f56fcf726cf4071132c8b779ad2eae0",
        "image_tag": "tensorfold-dsv41:fd2365042f56",
        "image_digest": null,
        "draft_tokens": 5,
        "configured_max_requests": 4,
        "context_limit_tokens": 1048576,
        "prefill_chunk_tokens": 2048,
        "chat_template": "patched kit template; source infers per-run use from file times",
        "workload": "code",
        "per_prompt_decode_rates": [
          91.1,
          92.5
        ],
        "per_prompt_TTFT_seconds": [
          0.13,
          0.16
        ],
        "prompt_count": 2
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-17/11_tensorfold-fd2365042f56_suite-results_32of32.json#L20",
      "source_document_sha256": "d0ec8c99b0208ac4ddc272379d3cd71c796b5662d09481d6ba1e98fbbf0fa41c",
      "source_document_id": "ds1017-runs-2026-10-17-11_tensorfold-fd2365042f56_suite-results_32of32.json",
      "source_line": 20,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    },
    {
      "label": "TensorFold final setup 32K prefill",
      "metric": "prefill_rate",
      "value": 2165,
      "unit": "tokens/s",
      "statistic": "median of five source rows; individually recoverable in per_run_values",
      "uncertainty": {
        "observed_min": 2065,
        "observed_max": 2201,
        "interval_kind": "observed range, not confidence interval"
      },
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-18",
        "concurrency": 1,
        "thinking": false,
        "temperature": null,
        "top_p": null,
        "max_output_tokens": null,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": false,
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "prompt_tokens / TTFT; source final setup and host tuning",
        "repetitions": 5,
        "checkpoint_is_latest_revision_verified": false,
        "engine": "TensorFold",
        "engine_version": "0.6.6",
        "runtime_execution_revision": null,
        "runtime_revision_prefix": "92aa710b533d",
        "image_tag": "tensorfold-dsv41:92aa710b533d",
        "image_digest": null,
        "reported_prompt_bucket": "32K",
        "cached_tokens": 0,
        "per_run_values": [
          2201,
          2180,
          2065,
          2165,
          2100
        ],
        "run_ids": [
          "hw5",
          "hw6",
          "ht1",
          "ht2",
          "ht3"
        ],
        "draft_tokens": 5,
        "context_limit_tokens": 1048576,
        "warmup": "one 8K prompt; subsequent 8K requests before 32K/64K/128K; random-word prompts",
        "host_tuning_scope": "author applies no engine swap and memory-reclaim/compaction controls; not a universal host-tuning recommendation"
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-18/INDEX.md#L16",
      "source_document_sha256": "e69d8c270e6407ddbaacb82da36cc48994b6f20fd4300d176753371b4e23fb8e",
      "source_document_id": "ds1017-runs-2026-10-18-INDEX.md",
      "source_line": 16,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    },
    {
      "label": "TensorFold final setup 128K prefill",
      "metric": "prefill_rate",
      "value": 1961,
      "unit": "tokens/s",
      "statistic": "median of five source rows; individually recoverable in per_run_values",
      "uncertainty": {
        "observed_min": 1851,
        "observed_max": 1986,
        "interval_kind": "observed range, not confidence interval"
      },
      "conditions": {
        "model": "DeepSeek-V4.1-Flash-1017",
        "model_artifact": "LibertAIDAI/DeepSeek-V4.1-Flash-1017-EXL3-2.9bpw",
        "model_revision": "afc7aa0aab98ee735c0f54e0a122019af1ac08f1",
        "Engram_artifact": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
        "Engram_revision": "1bae0e1088993b9b82de08f59de5be5a34e42170",
        "weight_precision": "EXL3 2.9bpw",
        "system": "2 x NVIDIA DGX Spark GB10, TP2 over ConnectX-7",
        "node_count": 2,
        "driver": "580.178.04",
        "CUDA": "13.0",
        "experiment_date": "2026-10-18",
        "concurrency": 1,
        "thinking": false,
        "temperature": null,
        "top_p": null,
        "max_output_tokens": null,
        "output_tokens": null,
        "input_tokens": null,
        "prefix_cache": false,
        "kv_dtype": null,
        "kv_pool_tokens": null,
        "measurement_denominator": "prompt_tokens / TTFT; source final setup and host tuning",
        "repetitions": 5,
        "checkpoint_is_latest_revision_verified": false,
        "engine": "TensorFold",
        "engine_version": "0.6.6",
        "runtime_execution_revision": null,
        "runtime_revision_prefix": "92aa710b533d",
        "image_tag": "tensorfold-dsv41:92aa710b533d",
        "image_digest": null,
        "reported_prompt_bucket": "128K",
        "cached_tokens": 0,
        "per_run_values": [
          1986,
          1981,
          1914,
          1961,
          1851
        ],
        "run_ids": [
          "hw5",
          "hw6",
          "ht1",
          "ht2",
          "ht3"
        ],
        "draft_tokens": 5,
        "context_limit_tokens": 1048576,
        "warmup": "one 8K prompt; subsequent 8K requests before 32K/64K/128K; random-word prompts",
        "host_tuning_scope": "author applies no engine swap and memory-reclaim/compaction controls; not a universal host-tuning recommendation"
      },
      "evidence_state": "source_reported",
      "site_reproduced": false,
      "evidence_url": "https://github.com/Elevatormusic/DeepSeek-V4.1-Flash-1017-EXL3-TensorFold-2x-DGX-Spark/blob/9c6ab02340b770c6c3e1a9a4d5e70a56c5829d32/runs/2026-10-18/INDEX.md#L16",
      "source_document_sha256": "e69d8c270e6407ddbaacb82da36cc48994b6f20fd4300d176753371b4e23fb8e",
      "source_document_id": "ds1017-runs-2026-10-18-INDEX.md",
      "source_line": 16,
      "source_record_id": "deepseek-1017-tensorfold-spark"
    }
  ],
  "conditions": {
    "source_reported": true,
    "independent_site_runs": 0,
    "same_pair_and_model_view_declared": true,
    "controlled_engine_only_AB": false,
    "old_six_build_median_is_not_repetition": true,
    "current_recipe_is_not_run11": true,
    "vision_tested": false,
    "perplexity_or_general_accuracy_tested": false
  },
  "deployment_profiles": [
    {
      "id": "vllm-kit-run01",
      "engine": "vLLM + exllamav3 overlay",
      "engine_version": null,
      "runtime_execution_revision": null,
      "image_digest": "sha256:2f0cf3adc0f989c1d446be274df864eb799630175f604c3b22b71b7205971dce",
      "kit_revision": "6f7d1590ad49a2b8995188e45d7b9db31e677452",
      "draft_tokens": 3,
      "configured_max_requests": 1,
      "context_limit_tokens": 262144,
      "prefill_chunk_tokens": 1536,
      "chat_template": "kit template; actual CHAT_TEMPLATE override not read by the source"
    },
    {
      "id": "tensorfold-run11",
      "engine": "TensorFold",
      "engine_version": "0.6.6",
      "runtime_execution_revision": "fd2365042f56fcf726cf4071132c8b779ad2eae0",
      "image_tag": "tensorfold-dsv41:fd2365042f56",
      "image_digest": null,
      "draft_tokens": 5,
      "configured_max_requests": 4,
      "context_limit_tokens": 1048576,
      "prefill_chunk_tokens": 2048,
      "chat_template": "patched kit template; source infers per-run use from file times"
    }
  ]
}
```