models / MiniMaxAI/MiniMax-H3
33B paramsother
Sparks
~74 GB
~41 GB
~25 GB
~18 GB
110 usable
Spec decode
On the Sparks — 1× · 1-bit GGUF (dynamic)
spark
93 GB free
110 usable
18* / 110 GB
16% of usable
Dense weightsActivations + graphsOS reserve
* segment sizes marked with an asterisk are estimates pending a measured run
Estimated per-node footprint, weights + KV sharded across the Spark (tensor parallel). Overhead ~8 GB/node; KV at FP16.
Optimize
Verdict
33B params, ~66 GB at native precision — fits on one Spark as-is.
Make it fast
Once it fits: fewer bytes per token and fewer decode steps against 273 GB/s.
Add a speculative decoder
Decode is bandwidth-bound at ~4.1 tok/s ceiling; a trained EAGLE-style draft multiplies tokens per weight-read.
Make it yours
Change what the model does — adapt, edit, or steer it — independent of size.
Fine-tune it
LoRA/QLoRA adapts behavior on one Spark without touching the base weights.
Edit or steer it
One-shot weight math — no gradients, no corpus.