# Is the DGX Spark worth it for local LLM work? A measured answer

> The highest-volume buyer question in this niche is answered almost entirely by spec-sheet arithmetic. Here is what a pair of GB10 nodes actually did in our lab — serving a 284B MoE, walking a context curve to a million tokens — and who should and should not buy one.

- Published: 2026-08-16
- Evidence level: lab_single_run
- Funding: Self-funded internal research
- Canonical: https://agmind.ai/reports/dgx-spark-worth-it/

Ask the web whether the DGX Spark is worth buying for local LLM work and you
get spec sheets: 128 GB unified, a petaflop of sparse FP4, ConnectX-7. What
you almost never get is what the box *did* for anyone under a real workload.
We own a pair, we published the raw runs, and this page answers the buyer
question only from what we measured — with the limits stated, because our
DGX lane is bring-up work (`lab_single_run`), not our three-repeat
methodology.

## What a pair of Sparks actually did

Everything below is quoted from our public run repositories and is
reproducible from the recipes in them.

**It served a 284B-parameter MoE at home.** DeepSeek-V4-Flash-0731 (284B
total, ~13B active), tensor-parallel across two nodes over 200G RoCE
. Single-stream client decode on the final configuration:
67.6 tok/s on a card-like short-output profile, 43.7 tok/s on a heavy
4K-context prose profile . That profile spread — not the hardware —
is the main reason published Spark numbers disagree with each other; the
full reconciliation is [its own page](https://agmind.ai/reports/dspark-speed-numbers-disagree/).

**It aggregated for a group.** At 12 concurrent streams the same
configuration produced 260.4 tok/s aggregate on the code profile
; in the earlier 1M-context build, 8 streams held ~99 tok/s
aggregate and 12 streams 141–146 tok/s . Aggregate is a serving
number, not a feel number: divide by your concurrency before imagining a
chat session.

**It walked a context curve to a million tokens — and told the truth about
it.** Decode ran 42.7 tok/s at 16K depth, 17.6 at 512K, 17.2 at 1.03M
— a plateau past the half-million mark, which is genuinely
remarkable. The bill arrives elsewhere: the *first* fill of a 1M-token
context cost 642 seconds cold, 4 seconds warm . A million-token
window you can afford to fill once is a different product from one you can
afford to fill per request.

**One flag mattered more than any tuning.** The runtime ships a MoE kernel
built for this model's MXFP4 experts and excludes it from auto-selection:
`--moe-backend flashinfer_b12x` moved single-stream decode from 59.7 to
67.6 tok/s client-side , a 9–12% step-rate gain measured
engine-side . The [backend report](https://agmind.ai/reports/dsv4-flash-moe-backend-dgx-spark/)
covers it; the [recipe page](https://agmind.ai/reports/deepseek-v4-flash-0731-dgx-spark-recipe/)
shows how to verify the flag actually took effect.

## Who the pair is for

- **You want the biggest open MoE class running under your own roof.** Two
  Sparks did that, stably enough to publish the runs. Nothing else at this
  power and desk footprint did, in our fleet.
- **You serve a small group, not one impatient person.** The aggregate
  numbers are real; the per-stream numbers at high concurrency are modest.
- **You do long-context research.** The depth plateau is a finding worth
  owning hardware for — if your sessions reuse the prefix, so the 642-second
   cold fill amortizes.

## Who should not buy it

- **Single-user, latency-sensitive work on heavy prompts.** 43.7 tok/s
  on the heavy profile is workable but not snappy, and every fresh
  long document pays prefill in full.
- **Anyone expecting datacenter Blackwell behavior.** Our
  [testbed page](https://agmind.ai/testbed/) states the limit plainly: GB10 results do not
  generalize upward.
- **Anyone unwilling to operate it.** Bring-up was real work — a one-line
  port bug ate days ([the saga](https://agmind.ai/reports/dspark-speculative-1m-dgx-spark/)),
  and standard GPU monitoring half-works: on GB10 unified memory,
  `dcgm-exporter` does not function and NVML returns N/A on a large share of
  queries; we ended up parsing `nvidia-smi` into a custom textfile collector
  ([write-up, in Russian](https://habr.com/ru/articles/1030802/)).

## Limits of this answer

One pair of units, one model family at the center, single passes with
warmups rather than our three-repeat discipline — `lab_single_run` is the
evidence level and the label means it. No price advice: hardware prices move
and this lab does not publish them. When our methodology-v1 lane reaches the
Sparks, the numbers on this page will be superseded by claims with
confidence intervals — subscribe on [/subscribe/](https://agmind.ai/subscribe/) if you want
that event, not this snapshot.

Raw runs, manifests and the serving recipe:
[botAGI/dspark-0731-gb10](https://github.com/botAGI/dspark-0731-gb10) and
[the 1M-context repository](https://github.com/botAGI/DeepSeek-V4-Flash-DSpark-GB10-2x-DGX-Spark-1m-fp4-fp8).

---

Machine-readable claim registry: https://agmind.ai/claims.json · llms.txt: https://agmind.ai/llms.txt