THE SOURCE IN CONTEXT
Original M3 weights and inference guidance, linking several local engines. This entry identifies the upstream model rather than a Spark-specific conversion.
Why does the model-size number differ between sources?
The official card gives approximately 428B total and 23B activated parameters, whereas the vLLM recipe lists 427B and 26B. The difference is unresolved in these references. We preserve both source declarations instead of averaging them or silently choosing one. The card advertises a native multimodal model with a 1M context window.
Source: MiniMaxAI · pinned document
What does the sparse-attention speed claim mean?
MiniMax describes relative prefill and decode improvements over M2 at a one-million-token context. The card does not provide a matching workstation measurement for the reader’s hardware in that sentence. Treat it as a publisher comparison under its own conditions, and use the separately dated three-Spark operator report for an actual deployment account.
Source: MiniMaxAI · pinned document
Which sampling settings are recommended?
The card recommends temperature 1.0 and top_p 0.95. Those recommendations differ from a greedy regression test at temperature zero. The three-Spark report contains both settings and reports different outcomes, so a reader should not combine those samples into a single accuracy or speed figure.
Source: MiniMaxAI · pinned document
How should a local serving route be chosen?
The publisher’s local-deployment section links engine integrations. The companion recipe defaults to a large BF16 artifact, while the three-Spark deployment uses an NVIDIA NVFP4 conversion, a DSpark drafter and custom tensor-parallel patches. Those are different artifacts and environments. Capture the conversion and runtime revisions before using either as a reproduction plan.
Source: MiniMaxAI · pinned document
What does the source specify?
These declarations belong to the cited document. A planning minimum, model limit and measured runtime allocation are different quantities.
| Card total / active parameters | approximately 428B / 23B |
|---|---|
| Published context | 1M tokens |
| Recommended sampling | temperature 1.0; top_p 0.95 |
AI-assisted source reading and original editorial synthesis; no independent model execution.