Skip to main content
fanout

Calculations run on this device. Scenario values are not sent anywhere — the engine is fully local.

inference calculator · free

Multi-GPU Parallelism Mapper

Map a model's layers and activations onto parallel strategies and compare memory.

Tensor, pipeline, or data parallel for this shape of model?

Engine 1.0.0 · calculation
164
2
2
2
TP×PP×DP
8
Matches GPU count?
yes
Unused / over
0
Comm tax (toy)
TP=2 PP=2

Parallelism factors tile the cluster.

2×2×2 vs 8 GPUs

Method

  • Deterministic local model of the mechanism — knobs recompute outputs instantly with no network.
  • Numbers are teaching approximations, not production capacity guarantees.

Sources

Keep going in inference

inference calculatorKV Cache CalculatorWhy do long context and batch size compete for the same GPU?inference calculatorGPU Model FitDoes this checkpoint actually fit on that GPU?inference calculatorLatency Numbers Every Engineer Should KnowWhich operation dominates this request's budget?