vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-11 17:44:38 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

Merge commit 'df270ef745' into concedo_experimental

Browse source 
# Conflicts:
#	Makefile
#	common/CMakeLists.txt
#	common/common.h
#	common/sampling.cpp
#	common/sampling.h
#	examples/infill/infill.cpp
#	examples/llama-bench/llama-bench.cpp
#	examples/quantize-stats/quantize-stats.cpp
#	examples/server/server.cpp
#	include/llama.h
#	src/llama-sampling.cpp
#	src/llama-sampling.h
#	src/llama.cpp
#	tests/test-grammar-integration.cpp
#	tests/test-grammar-parser.cpp
#	tests/test-json-schema-to-grammar.cpp
#	tests/test-llama-grammar.cpp
#	tests/test-sampling.cpp
This commit is contained in:
Concedo 2024-09-09 17:10:08 +08:00
commit 12fd16bfd4
86 changed files with 3406 additions and 7795 deletions
Download patch file Download diff file Expand all files Collapse all files

12
examples/llama.swiftui/README.md
View file

@ -1,12 +0,0 @@
# llama.cpp/examples/llama.swiftui
Local inference of llama.cpp on an iPhone. This is a sample app that can be used as a starting
point for more advanced projects.
For usage instructions and performance stats, check the following discussion: https://github.com/ggerganov/llama.cpp/discussions/4508
![image](https://github.com/ggerganov/llama.cpp/assets/1991296/2b40284f-8421-47a2-b634-74eece09a299)
Video demonstration:
https://github.com/bachittle/llama.cpp/assets/39804642/e290827a-4edb-4093-9642-2a5e399ec545

23
examples/llama.swiftui/llama.cpp.swift/LibLlama.swift
View file

@ -24,6 +24,7 @@ func llama_batch_add(_ batch: inout llama_batch, _ id: llama_token, _ pos: llama
actor LlamaContext {
private var model: OpaquePointer
private var context: OpaquePointer
private var sampling: UnsafeMutablePointer<llama_sampler>
private var batch: llama_batch
private var tokens_list: [llama_token]
var is_done: Bool = false
@ -42,9 +43,15 @@ actor LlamaContext {
self.tokens_list = []
self.batch = llama_batch_init(512, 0, 1)
self.temporary_invalid_cchars = []
let sparams = llama_sampler_chain_default_params()
self.sampling = llama_sampler_chain_init(sparams)
llama_sampler_chain_add(self.sampling, llama_sampler_init_temp(0.4))
llama_sampler_chain_add(self.sampling, llama_sampler_init_softmax())
llama_sampler_chain_add(self.sampling, llama_sampler_init_dist(1234))
}
deinit {
llama_sampler_free(sampling)
llama_batch_free(batch)
llama_free(context)
llama_free_model(model)
@ -69,7 +76,6 @@ actor LlamaContext {
print("Using \(n_threads) threads")
var ctx_params = llama_context_default_params()
ctx_params.seed = 1234
ctx_params.n_ctx = 2048
ctx_params.n_threads = Int32(n_threads)
ctx_params.n_threads_batch = Int32(n_threads)
@ -144,20 +150,9 @@ actor LlamaContext {
func completion_loop() -> String {
var new_token_id: llama_token = 0
let n_vocab = llama_n_vocab(model)
let logits = llama_get_logits_ith(context, batch.n_tokens - 1)
new_token_id = llama_sampler_sample(sampling, context, batch.n_tokens - 1)
var candidates = Array<llama_token_data>()
candidates.reserveCapacity(Int(n_vocab))
for token_id in 0..<n_vocab {
candidates.append(llama_token_data(id: token_id, logit: logits![Int(token_id)], p: 0.0))
}
candidates.withUnsafeMutableBufferPointer() { buffer in
var candidates_p = llama_token_data_array(data: buffer.baseAddress, size: buffer.count, sorted: false)
new_token_id = llama_sample_token_greedy(context, &candidates_p)
}
llama_sampler_accept(sampling, new_token_id)
if llama_token_is_eog(model, new_token_id) || n_cur == n_len {
print("\n")