For the complete documentation index, see llms.txt. Markdown versions of all pages are available by appending .md to any URL (e.g. /get-started.md).

Python module

max.nn.sampling

Sampling custom ops.

Acceptance sampling

`AcceptanceSampler`	Dispatches between greedy, synthetic, and stochastic acceptance.

`compute_synthetic_acceptance_base_rate`	Solves for the per-position base acceptance rate that matches a target mean.
`greedy_acceptance_sampler`	Target-only rejection sampler for speculative decoding.
`stochastic_acceptance_sampler`	Verifies speculative draft tokens against the target model.
`synthetic_acceptance_sampler`	Synthetic sampler for speculative decoding benchmarking.