Files
llama.cpp/src/llama-sampler.h
T
Gaurav GargandGeorgi Gerganov dd1ea52433 llama : support multi-output backend sampling (#25532)
* Enable backend sampling with token speculation

* Clamp the mask sum before converting it into the sampled index

* Add a numeric context parameter declaring the maximum outputs one sequence

* More fixes

* Don't reuse memory for output views.

* Match dist between CPU and GPU

* Fix CPU and backend sampling mismatches

* Simpify some of the changes

* Fix tests on Vulkan

* More test fixes

* Rebase changes

* Rebase and address review comments

* Address review comments

* Address review comments

* Update src/llama-sampler.cpp

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>

---------

Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
2026-08-10 16:58:56 +03:00

47 lines
982 B
C++

#pragma once
#include "llama.h"
#include <vector>
struct llama_vocab;
struct llama_grammar;
// sampler chain
struct llama_sampler_chain {
llama_sampler_chain_params params;
// has .backend_init() been called?
bool is_init = false;
uint32_t n_nodes = 0;
struct info {
bool is_backend;
llama_sampler * ptr;
};
std::vector<info> samplers;
// pre-allocated buffer for llama_sampler_sample to avoid repeated allocations
std::vector<llama_token_data> cur;
// timing
mutable int64_t t_sample_us;
mutable int32_t n_sample;
};
uint32_t llama_sampler_backend_n_nodes(const llama_sampler * sampler);
void llama_sampler_backend_begin(llama_sampler * sampler);
struct llama_sampler * llama_sampler_init_dry_testing(
float dry_multiplier,
float dry_base,
int32_t dry_allowed_length,
int32_t dry_penalty_last_n,
const std::vector<std::vector<llama_token>> & seq_breakers);