vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-13 02:19:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Merge branch 'upstream' into concedo_experimental

Browse source 
# Conflicts:
#	docs/backend/CANN.md
#	docs/multimodal/minicpmo2.6.md
#	docs/multimodal/minicpmv2.5.md
#	docs/multimodal/minicpmv2.6.md
#	examples/speculative-simple/speculative-simple.cpp
#	ggml/cmake/ggml-config.cmake.in
#	ggml/src/ggml-cann/aclnn_ops.cpp
#	ggml/src/ggml-cann/ggml-cann.cpp
#	ggml/src/ggml-cpu/repack.cpp
#	ggml/src/ggml-opencl/CMakeLists.txt
#	ggml/src/ggml-opencl/ggml-opencl.cpp
#	ggml/src/ggml-opencl/kernels/add.cl
#	ggml/src/ggml-opencl/kernels/mul.cl
#	scripts/compare-commits.sh
#	scripts/compare-llama-bench.py
#	scripts/sync-ggml.last
#	tools/server/README.md
This commit is contained in:
Concedo 2025-08-02 10:25:10 +08:00
commit f430916a71
57 changed files with 6028 additions and 731 deletions
Download patch file Download diff file Expand all files Collapse all files

78
common/arg.cpp
View file

@ -979,6 +979,10 @@ static bool common_params_parse_ex(int argc, char ** argv, common_params_context
for (auto & seq_breaker : params.sampling.dry_sequence_breakers) {
string_process_escapes(seq_breaker);
}
for (auto & pair : params.speculative.replacements) {
string_process_escapes(pair.first);
string_process_escapes(pair.second);
}
}
if (!params.kv_overrides.empty()) {
@ -2093,6 +2097,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.no_kv_offload = true;
}
).set_env("LLAMA_ARG_NO_KV_OFFLOAD"));
add_opt(common_arg(
{"-nr", "--no-repack"},
"disable weight repacking",
[](common_params & params) {
params.no_extra_bufts = true;
}
).set_env("LLAMA_ARG_NO_REPACK"));
add_opt(common_arg(
{"-ctk", "--cache-type-k"}, "TYPE",
string_format(
@ -2371,6 +2382,15 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
}
));
add_opt(common_arg(
{"--cpu-moe"},
"use CPU for Mixture of Experts (MoE) weights",
[](common_params & params) {
params.tensor_buft_overrides.push_back({"\\.ffn_up_exps\\.weight$", ggml_backend_cpu_buffer_type()});
params.tensor_buft_overrides.push_back({"\\.ffn_down_exps\\.weight$", ggml_backend_cpu_buffer_type()});
params.tensor_buft_overrides.push_back({"\\.ffn_gate_exps\\.weight$", ggml_backend_cpu_buffer_type()});
}
).set_env("LLAMA_ARG_CPU_MOE"));
add_opt(common_arg(
{"-ngl", "--gpu-layers", "--n-gpu-layers"}, "N",
"number of layers to store in VRAM",
@ -3251,6 +3271,13 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
params.speculative.model.path = value;
}
).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}).set_env("LLAMA_ARG_MODEL_DRAFT"));
add_opt(common_arg(
{"--spec-replace"}, "TARGET", "DRAFT",
"translate the string in TARGET into DRAFT if the draft model and main model are not compatible",
[](common_params & params, const std::string & tgt, const std::string & dft) {
params.speculative.replacements.push_back({ tgt, dft });
}
).set_examples({LLAMA_EXAMPLE_SPECULATIVE, LLAMA_EXAMPLE_SERVER}));
add_opt(common_arg(
{"-ctkd", "--cache-type-k-draft"}, "TYPE",
string_format(
@ -3440,28 +3467,11 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
}
).set_examples({LLAMA_EXAMPLE_SERVER}));
// diffusion parameters
add_opt(common_arg(
{ "--diffusion-steps" }, "N",
string_format("number of diffusion steps (default: %d)", params.diffusion.steps),
[](common_params & params, int value) { params.diffusion.steps = value; }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-eps" }, "F",
string_format("epsilon for timesteps (default: %.6f)", (double) params.diffusion.eps),
[](common_params & params, const std::string & value) { params.diffusion.eps = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-algorithm" }, "N",
string_format("diffusion algorithm: 0=ORIGIN, 1=MASKGIT_PLUS, 2=TOPK_MARGIN, 3=ENTROPY (default: %d)",
params.diffusion.algorithm),
[](common_params & params, int value) { params.diffusion.algorithm = value; }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-alg-temp" }, "F",
string_format("algorithm temperature (default: %.3f)", (double) params.diffusion.alg_temp),
[](common_params & params, const std::string & value) { params.diffusion.alg_temp = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-visual" },
string_format("enable visual diffusion mode (show progressive generation) (default: %s)",
@ -3469,5 +3479,39 @@ common_params_context common_params_parser_init(common_params & params, llama_ex
[](common_params & params) { params.diffusion.visual_mode = true; }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-eps" }, "F",
string_format("epsilon for timesteps (default: %.6f)", (double) params.diffusion.eps),
[](common_params & params, const std::string & value) { params.diffusion.eps = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-algorithm" }, "N",
string_format("diffusion algorithm: 0=ORIGIN, 1=ENTROPY_BASED, 2=MARGIN_BASED, 3=RANDOM, 4=LOW_CONFIDENCE (default: %d)",
params.diffusion.algorithm),
[](common_params & params, int value) { params.diffusion.algorithm = value; }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-alg-temp" }, "F",
string_format("dream algorithm temperature (default: %.3f)", (double) params.diffusion.alg_temp),
[](common_params & params, const std::string & value) { params.diffusion.alg_temp = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-block-length" }, "N",
string_format("llada block length for generation (default: %d)", params.diffusion.block_length),
[](common_params & params, int value) { params.diffusion.block_length = value; }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-cfg-scale" }, "F",
string_format("llada classifier-free guidance scale (default: %.3f)", (double) params.diffusion.cfg_scale),
[](common_params & params, const std::string & value) { params.diffusion.cfg_scale = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
add_opt(common_arg(
{ "--diffusion-add-gumbel-noise" }, "F",
string_format("add gumbel noise to the logits if temp > 0.0 (default: %s)", params.diffusion.add_gumbel_noise ? "true" : "false"),
[](common_params & params, const std::string & value) { params.diffusion.add_gumbel_noise = std::stof(value); }
).set_examples({ LLAMA_EXAMPLE_DIFFUSION }));
return ctx_arg;
}